{
  "session_id": "20260530_094622_159128",
  "model": "mimo-v2.5-pro",
  "base_url": "https://api.xiaomimimo.com/v1",
  "platform": "curator",
  "session_start": "2026-05-30T09:46:22.143424",
  "last_updated": "2026-05-30T09:47:11.656223",
  "system_prompt": "You are Hermes Agent, an intelligent AI assistant created by Nous Research. You are helpful, knowledgeable, and direct. You assist users with a wide range of tasks including answering questions, writing and editing code, analyzing information, creative work, and executing actions via your tools. You communicate clearly, admit uncertainty when appropriate, and prioritize being genuinely useful over being verbose unless otherwise directed below. Be targeted and efficient in your exploration and investigations.\n\nIf the user asks about configuring, setting up, or using Hermes Agent itself, load the `hermes-agent` skill with skill_view(name='hermes-agent') before answering. Docs: https://hermes-agent.nousresearch.com/docs\n\nYou have persistent memory across sessions. Save durable facts using the memory tool: user preferences, environment details, tool quirks, and stable conventions. Memory is injected into every turn, so keep it compact and focused on facts that will still matter later.\nPrioritize what reduces future user steering — the most valuable memory is one that prevents the user from having to correct or remind you again. User preferences and recurring corrections matter more than procedural task details.\nDo NOT save task progress, session outcomes, completed-work logs, or temporary TODO state to memory; use session_search to recall those from past transcripts. If you've discovered a new way to do something, solved a problem that could be necessary later, save it as a skill with the skill tool.\nWrite memories as declarative facts, not instructions to yourself. 'User prefers concise responses' ✓ — 'Always respond concisely' ✗. 'Project uses pytest with xdist' ✓ — 'Run tests with pytest -n 4' ✗. Imperative phrasing gets re-read as a directive in later sessions and can cause repeated work or override the user's current request. Procedures and workflows belong in skills, not memory. When the user references something from a past conversation or you suspect relevant cross-session context exists, use session_search to recall it before asking them to repeat themselves. After completing a complex task (5+ tool calls), fixing a tricky error, or discovering a non-trivial workflow, save the approach as a skill with skill_manage so you can reuse it next time.\nWhen using a skill and finding it outdated, incomplete, or wrong, patch it immediately with skill_manage(action='patch') — don't wait to be asked. Skills that aren't maintained become liabilities.\n\n## Skills (mandatory)\nBefore replying, scan the skills below. If a skill matches or is even partially relevant to your task, you MUST load it with skill_view(name) and follow its instructions. Err on the side of loading — it is always better to have context you don't need than to miss critical steps, pitfalls, or established workflows. Skills contain specialized knowledge — API endpoints, tool-specific commands, and proven workflows that outperform general-purpose approaches. Load the skill even if you think you could handle the task with basic tools like web_search or terminal. Skills also encode the user's preferred approach, conventions, and quality standards for tasks like code review, planning, and testing — load them even for tasks you already know how to do, because the skill defines how it should be done here.\nWhenever the user asks you to configure, set up, install, enable, disable, modify, or troubleshoot Hermes Agent itself — its CLI, config, models, providers, tools, skills, voice, gateway, plugins, or any feature — load the `hermes-agent` skill first. It has the actual commands (e.g. `hermes config set …`, `hermes tools`, `hermes setup`) so you don't have to guess or invent workarounds.\nIf a skill has issues, fix it with skill_manage(action='patch').\nAfter difficult/iterative tasks, offer to save as a skill. If a skill you loaded was missing steps, had wrong commands, or needed pitfalls you discovered, update it before finishing.\n\n<available_skills>\n  autonomous-ai-agents: Skills for spawning and orchestrating autonomous AI coding agents and multi-agent workflows — running independent agent processes, delegating tasks, and coordinating parallel workstreams.\n    - claude-code: Delegate coding to Claude Code CLI (features, PRs).\n    - codex: Delegate coding to OpenAI Codex CLI (features, PRs).\n    - hermes-agent: Configure, extend, or contribute to Hermes Agent.\n    - opencode: Delegate coding to OpenCode CLI (features, PR review).\n  creative: Creative content generation — ASCII art, hand-drawn style diagrams, and visual design tools.\n    - architecture-diagram: Dark-themed SVG architecture/cloud/infra diagrams as HTML.\n    - ascii-art: ASCII art: pyfiglet, cowsay, boxes, image-to-ascii.\n    - ascii-video: ASCII video: convert video/audio to colored ASCII MP4/GIF.\n    - baoyu-comic: Knowledge comics (知识漫画): educational, biography, tutorial.\n    - baoyu-infographic: Infographics: 21 layouts x 21 styles (信息图, 可视化).\n    - claude-design: Design one-off HTML artifacts (landing, deck, prototype).\n    - comfyui: Generate images, video, and audio with ComfyUI — install,...\n    - design-md: Author/validate/export Google's DESIGN.md token spec files.\n    - excalidraw: Hand-drawn Excalidraw JSON diagrams (arch, flow, seq).\n    - humanizer: Humanize text: strip AI-isms and add real voice.\n    - hyperframes-deploy: Deploy and use HyperFrames — HTML-to-video rendering fram...\n    - ideation: Generate project ideas via creative constraints.\n    - manim-video: Manim CE animations: 3Blue1Brown math/algo videos.\n    - p5js: p5.js sketches: gen art, shaders, interactive, 3D.\n    - pixel-art: Pixel art w/ era palettes (NES, Game Boy, PICO-8).\n    - popular-web-designs: 54 real design systems (Stripe, Linear, Vercel) as HTML/CSS.\n    - pretext: Use when building creative browser demos with @chenglou/p...\n    - sketch: Throwaway HTML mockups: 2-3 design variants to compare.\n    - songwriting-and-ai-music: Songwriting craft and Suno AI music prompts.\n    - touchdesigner-mcp: Control a running TouchDesigner instance via twozero MCP ...\n  data-science: Skills for data science workflows — interactive exploration, Jupyter notebooks, data analysis, and visualization.\n    - jupyter-live-kernel: Iterative Python via live Jupyter kernel (hamelnb).\n  devops:\n    - docker: Docker on Linux: install Engine + Compose, configure prox...\n    - kanban-orchestrator: Decomposition playbook + specialist-roster conventions + ...\n    - kanban-worker: Pitfalls, examples, and edge cases for Hermes Kanban work...\n    - python-webapp-deploy: Deploy Python/Flask/Node.js web applications from source ...\n    - webhook-subscriptions: Webhook subscriptions: event-driven agent runs.\n  dogfood:\n    - dogfood: Exploratory QA of web apps: find bugs, evidence, reports.\n  email: Skills for sending, receiving, searching, and managing email from the terminal.\n    - himalaya: Himalaya CLI: IMAP/SMTP email from terminal.\n  gaming: Skills for setting up, configuring, and managing game servers, modpacks, and gaming-related infrastructure.\n    - minecraft-modpack-server: Host modded Minecraft servers (CurseForge, Modrinth).\n    - pokemon-player: Play Pokemon via headless emulator + RAM reads.\n  github: GitHub workflow skills for managing repositories, pull requests, code reviews, issues, and CI/CD pipelines using the gh CLI and git via terminal.\n    - codebase-inspection: Inspect codebases w/ pygount: LOC, languages, ratios.\n    - github-auth: GitHub auth setup: HTTPS tokens, SSH keys, gh CLI login.\n    - github-code-review: Review PRs: diffs, inline comments via gh or REST.\n    - github-issues: Create, triage, label, assign GitHub issues via gh or REST.\n    - github-pr-workflow: GitHub PR lifecycle: branch, commit, open, CI, merge.\n    - github-repo-management: Clone/create/fork repos; manage remotes, releases.\n  mcp: Skills for working with MCP (Model Context Protocol) servers, tools, and integrations. Documents the built-in native MCP client — configure servers in config.yaml for automatic tool discovery.\n    - native-mcp: MCP client: connect servers, register tools (stdio/HTTP).\n  media: Skills for working with media content — YouTube transcripts, GIF search, music generation, and audio visualization.\n    - gif-search: Search/download GIFs from Tenor via curl + jq.\n    - heartmula: HeartMuLa: Suno-like song generation from lyrics + tags.\n    - songsee: Audio spectrograms/features (mel, chroma, MFCC) via CLI.\n    - spotify: Spotify: play, search, queue, manage playlists and devices.\n    - youtube-content: YouTube transcripts to summaries, threads, blogs.\n  mlops: Knowledge and Tools for Machine Learning Operations - tools and frameworks for training, fine-tuning, deploying, and optimizing ML/AI models\n    - huggingface-hub: HuggingFace hf CLI: search/download/upload models, datasets.\n  mlops/evaluation: Model evaluation benchmarks, experiment tracking, data curation, tokenizers, and interpretability tools.\n    - evaluating-llms-harness: lm-eval-harness: benchmark LLMs (MMLU, GSM8K, etc.).\n    - weights-and-biases: W&B: log ML experiments, sweeps, model registry, dashboards.\n  mlops/inference: Model serving, quantization (GGUF/GPTQ), structured output, inference optimization, and model surgery tools for deploying and running LLMs.\n    - llama-cpp: llama.cpp local GGUF inference + HF Hub model discovery.\n    - obliteratus: OBLITERATUS: abliterate LLM refusals (diff-in-means).\n    - outlines: Outlines: structured JSON/regex/Pydantic LLM generation.\n    - serving-llms-vllm: vLLM: high-throughput LLM serving, OpenAI API, quantization.\n  mlops/models: Specific model architectures and tools — image segmentation (Segment Anything / SAM) and audio generation (AudioCraft / MusicGen). Additional model skills (CLIP, Stable Diffusion, Whisper, LLaVA) are available as optional skills.\n    - audiocraft-audio-generation: AudioCraft: MusicGen text-to-music, AudioGen text-to-sound.\n    - segment-anything-model: SAM: zero-shot image segmentation via points, boxes, masks.\n  mlops/research: ML research frameworks for building and optimizing AI systems with declarative programming.\n    - dspy: DSPy: declarative LM programs, auto-optimize prompts, RAG.\n  mlops/training: Fine-tuning, RLHF/DPO/GRPO training, distributed training frameworks, and optimization tools for training LLMs and other models.\n    - axolotl: Axolotl: YAML LLM fine-tuning (LoRA, DPO, GRPO).\n    - fine-tuning-with-trl: TRL: SFT, DPO, PPO, GRPO, reward modeling for LLM RLHF.\n    - unsloth: Unsloth: 2-5x faster LoRA/QLoRA fine-tuning, less VRAM.\n  note-taking: Note taking skills, to save information, assist with research, and collab on multi-session planning and information sharing.\n    - obsidian: Read, search, and create notes in the Obsidian vault.\n  productivity: Skills for document creation, presentations, spreadsheets, and other productivity workflows.\n    - airtable: Airtable REST API via curl. Records CRUD, filters, upserts.\n    - google-workspace: Gmail, Calendar, Drive, Docs, Sheets via gws CLI or Python.\n    - linear: Linear: manage issues, projects, teams via GraphQL + curl.\n    - maps: Geocode, POIs, routes, timezones via OpenStreetMap/OSRM.\n    - nano-pdf: Edit PDF text/typos/titles via nano-pdf CLI (NL prompts).\n    - notion: Notion API via curl: pages, databases, blocks, search.\n    - ocr-and-documents: Extract text from PDFs/scans (pymupdf, marker-pdf).\n    - powerpoint: Create, read, edit .pptx decks, slides, notes, templates.\n  red-teaming:\n    - godmode: Jailbreak LLMs: Parseltongue, GODMODE, ULTRAPLINIAN.\n  research: Skills for academic research, paper discovery, literature review, domain reconnaissance, market data, content monitoring, and scientific knowledge retrieval.\n    - arxiv: Search arXiv papers by keyword, author, category, or ID.\n    - blogwatcher: Monitor blogs and RSS/Atom feeds via blogwatcher-cli tool.\n    - llm-wiki: Karpathy's LLM Wiki: build/query interlinked markdown KB.\n    - polymarket: Query Polymarket: markets, prices, orderbooks, history.\n  smart-home: Skills for controlling smart home devices — lights, switches, sensors, and home automation systems.\n    - openhue: Control Philips Hue lights, scenes, rooms via OpenHue CLI.\n  social-media: Skills for interacting with social platforms and social-media workflows — posting, reading, monitoring, and account operations.\n    - xurl: X/Twitter via xurl CLI: post, search, DM, media, v2 API.\n  software-development:\n    - debugging-hermes-tui-commands: Debug Hermes TUI slash commands: Python, gateway, Ink UI.\n    - hermes-agent-skill-authoring: Author in-repo SKILL.md: frontmatter, validator, structure.\n    - node-inspect-debugger: Debug Node.js via --inspect + Chrome DevTools Protocol CLI.\n    - plan: Plan mode: write markdown plan to .hermes/plans/, no exec.\n    - python-debugpy: Debug Python: pdb REPL + debugpy remote (DAP).\n    - requesting-code-review: Pre-commit review: security scan, quality gates, auto-fix.\n    - spike: Throwaway experiments to validate an idea before build.\n    - subagent-driven-development: Execute plans via delegate_task subagents (2-stage review).\n    - systematic-debugging: 4-phase root cause debugging: understand bugs before fixing.\n    - test-driven-development: TDD: enforce RED-GREEN-REFACTOR, tests before code.\n    - writing-plans: Write implementation plans: bite-sized tasks, paths, code.\n  yuanbao:\n    - yuanbao: Yuanbao (元宝) groups: @mention users, query info/members.\n</available_skills>\n\nOnly proceed without loading a skill if genuinely none are relevant to the task.\n\nConversation started: Saturday, May 30, 2026 09:46 AM\nModel: mimo-v2.5-pro\nProvider: xiaomi",
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "clarify",
        "description": "Ask the user a question when you need clarification, feedback, or a decision before proceeding. Supports two modes:\n\n1. **Multiple choice** — provide up to 4 choices. The user picks one or types their own answer via a 5th 'Other' option.\n2. **Open-ended** — omit choices entirely. The user types a free-form response.\n\nUse this tool when:\n- The task is ambiguous and you need the user to choose an approach\n- You want post-task feedback ('How did that work out?')\n- You want to offer to save a skill or update memory\n- A decision has meaningful trade-offs the user should weigh in on\n\nDo NOT use this tool for simple yes/no confirmation of dangerous commands (the terminal tool handles that). Prefer making a reasonable default choice yourself when the decision is low-stakes.",
        "parameters": {
          "type": "object",
          "properties": {
            "question": {
              "type": "string",
              "description": "The question to present to the user."
            },
            "choices": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "maxItems": 4,
              "description": "Up to 4 answer choices. Omit this parameter entirely to ask an open-ended question. When provided, the UI automatically appends an 'Other (type your answer)' option."
            }
          },
          "required": [
            "question"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "cronjob",
        "description": "Manage scheduled cron jobs with a single compressed tool.\n\nUse action='create' to schedule a new job from a prompt or one or more skills.\nUse action='list' to inspect jobs.\nUse action='update', 'pause', 'resume', 'remove', or 'run' to manage an existing job.\n\nTo stop a job the user no longer wants: first action='list' to find the job_id, then action='remove' with that job_id. Never guess job IDs — always list first.\n\nJobs run in a fresh session with no current-chat context, so prompts must be self-contained.\nIf skills are provided on create, the future cron run loads those skills in order, then follows the prompt as the task instruction.\nOn update, passing skills=[] clears attached skills.\n\nNOTE: The agent's final response is auto-delivered to the target. Put the primary\nuser-facing content in the final response. Cron jobs run autonomously with no user\npresent — they cannot ask questions or request clarification.\n\nImportant safety rule: cron-run sessions should not recursively schedule more cron jobs.",
        "parameters": {
          "type": "object",
          "properties": {
            "action": {
              "type": "string",
              "description": "One of: create, list, update, pause, resume, remove, run"
            },
            "job_id": {
              "type": "string",
              "description": "Required for update/pause/resume/remove/run"
            },
            "prompt": {
              "type": "string",
              "description": "For create: the full self-contained prompt. If skills are also provided, this becomes the task instruction paired with those skills."
            },
            "schedule": {
              "type": "string",
              "description": "For create/update: '30m', 'every 2h', '0 9 * * *', or ISO timestamp"
            },
            "name": {
              "type": "string",
              "description": "Optional human-friendly name"
            },
            "repeat": {
              "type": "integer",
              "description": "Optional repeat count. Omit for defaults (once for one-shot, forever for recurring)."
            },
            "deliver": {
              "type": "string",
              "description": "Omit this parameter to auto-deliver back to the current chat and topic (recommended). Auto-detection preserves thread/topic context. Only set explicitly when the user asks to deliver somewhere OTHER than the current conversation. Values: 'origin' (same as omitting), 'local' (no delivery, save only), or platform:chat_id:thread_id for a specific destination. Examples: 'telegram:-1001234567890:17585', 'discord:#engineering', 'sms:+15551234567'. WARNING: 'platform:chat_id' without :thread_id loses topic targeting."
            },
            "skills": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "description": "Optional ordered list of skill names to load before executing the cron prompt. On update, pass an empty array to clear attached skills."
            },
            "model": {
              "type": "object",
              "description": "Optional per-job model override. If provider is omitted, the current main provider is pinned at creation time so the job stays stable.",
              "properties": {
                "provider": {
                  "type": "string",
                  "description": "Provider name (e.g. 'openrouter', 'anthropic'). Omit to use and pin the current provider."
                },
                "model": {
                  "type": "string",
                  "description": "Model name (e.g. 'anthropic/claude-sonnet-4', 'claude-sonnet-4')"
                }
              },
              "required": [
                "model"
              ]
            },
            "script": {
              "type": "string",
              "description": "Optional path to a Python script that runs before each cron job execution. Its stdout is injected into the prompt as context. Use for data collection and change detection. Relative paths resolve under ~/.hermes/scripts/. On update, pass empty string to clear."
            },
            "context_from": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "description": "Optional job ID or list of job IDs whose most recent completed output is injected into the prompt as context before each run. Use this to chain cron jobs: job A collects data, job B processes it. Each entry must be a valid job ID (from cronjob action='list'). Note: injects the most recent completed output — does not wait for upstream jobs running in the same tick. On update, pass an empty array to clear."
            },
            "enabled_toolsets": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "description": "Optional list of toolset names to restrict the job's agent to (e.g. [\"web\", \"terminal\", \"file\", \"delegation\"]). When set, only tools from these toolsets are loaded, significantly reducing input token overhead. When omitted, all default tools are loaded. Infer from the job's prompt — e.g. use \"web\" if it calls web_search, \"terminal\" if it runs scripts, \"file\" if it reads files, \"delegation\" if it calls delegate_task. On update, pass an empty array to clear."
            },
            "workdir": {
              "type": "string",
              "description": "Optional absolute path to run the job from. When set, AGENTS.md / CLAUDE.md / .cursorrules from that directory are injected into the system prompt, and the terminal/file/code_exec tools use it as their working directory — useful for running a job inside a specific project repo. Must be an absolute path that exists. When unset (default), preserves the original behaviour: no project context files, tools use the scheduler's cwd. On update, pass an empty string to clear. Jobs with workdir run sequentially (not parallel) to keep per-job directories isolated."
            }
          },
          "required": [
            "action"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "delegate_task",
        "description": "Spawn one or more subagents to work on tasks in isolated contexts. Each subagent gets its own conversation, terminal session, and toolset. Only the final summary is returned -- intermediate tool results never enter your context window.\n\nTWO MODES (one of 'goal' or 'tasks' is required):\n1. Single task: provide 'goal' (+ optional context, toolsets)\n2. Batch (parallel): provide 'tasks' array with up to delegation.max_concurrent_children items (default 3, configurable via config.yaml, no hard ceiling). All run concurrently and results are returned together. Nested delegation requires role='orchestrator' and delegation.max_spawn_depth >= 2.\n\nWHEN TO USE delegate_task:\n- Reasoning-heavy subtasks (debugging, code review, research synthesis)\n- Tasks that would flood your context with intermediate data\n- Parallel independent workstreams (research A and B simultaneously)\n\nWHEN NOT TO USE (use these instead):\n- Mechanical multi-step work with no reasoning needed -> use execute_code\n- Single tool call -> just call the tool directly\n- Tasks needing user interaction -> subagents cannot use clarify\n- Durable long-running work that must outlive the current turn -> use cronjob (action='create') or terminal(background=True, notify_on_complete=True) instead. delegate_task runs SYNCHRONOUSLY inside the parent turn: if the parent is interrupted (user sends a new message, /stop, /new) the child is cancelled with status='interrupted' and its work is discarded. Children cannot continue in the background.\n\nIMPORTANT:\n- Subagents have NO memory of your conversation. Pass all relevant info (file paths, error messages, constraints) via the 'context' field.\n- If the user is writing in a non-English language, or asked for output in a specific language / tone / style, say so in 'context' (e.g. \"respond in Chinese\", \"return output in Japanese\"). Otherwise subagents default to English and their summaries will contaminate your final reply with the wrong language.\n- Subagent summaries are SELF-REPORTS, not verified facts. A subagent that claims \"uploaded successfully\" or \"file written\" may be wrong. For operations with external side-effects (HTTP POST/PUT, remote writes, file creation at shared paths, publishing), require the subagent to return a verifiable handle (URL, ID, absolute path, HTTP status) and verify it yourself — fetch the URL, stat the file, read back the content — before telling the user the operation succeeded.\n- Leaf subagents (role='leaf', the default) CANNOT call: delegate_task, clarify, memory, send_message, execute_code.\n- Orchestrator subagents (role='orchestrator') retain delegate_task so they can spawn their own workers, but still cannot use clarify, memory, send_message, or execute_code. Orchestrators are bounded by delegation.max_spawn_depth (default 2) and can be disabled globally via delegation.orchestrator_enabled=false.\n- Each subagent gets its own terminal session (separate working directory and state).\n- Results are always returned as an array, one entry per task.",
        "parameters": {
          "type": "object",
          "properties": {
            "goal": {
              "type": "string",
              "description": "What the subagent should accomplish. Be specific and self-contained -- the subagent knows nothing about your conversation history."
            },
            "context": {
              "type": "string",
              "description": "Background information the subagent needs: file paths, error messages, project structure, constraints. The more specific you are, the better the subagent performs."
            },
            "toolsets": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "description": "Toolsets to enable for this subagent. Default: inherits your enabled toolsets. Available toolsets: 'browser', 'cronjob', 'discord', 'discord_admin', 'feishu_doc', 'feishu_drive', 'file', 'homeassistant', 'image_gen', 'kanban', 'search', 'session_search', 'skills', 'spotify', 'terminal', 'todo', 'tts', 'vision', 'web', 'yuanbao'. Common patterns: ['terminal', 'file'] for code work, ['web'] for research, ['browser'] for web interaction, ['terminal', 'file', 'web'] for full-stack tasks."
            },
            "tasks": {
              "type": "array",
              "items": {
                "type": "object",
                "properties": {
                  "goal": {
                    "type": "string",
                    "description": "Task goal"
                  },
                  "context": {
                    "type": "string",
                    "description": "Task-specific context"
                  },
                  "toolsets": {
                    "type": "array",
                    "items": {
                      "type": "string"
                    },
                    "description": "Toolsets for this specific task. Available: 'browser', 'cronjob', 'discord', 'discord_admin', 'feishu_doc', 'feishu_drive', 'file', 'homeassistant', 'image_gen', 'kanban', 'search', 'session_search', 'skills', 'spotify', 'terminal', 'todo', 'tts', 'vision', 'web', 'yuanbao'. Use 'web' for network access, 'terminal' for shell, 'browser' for web interaction."
                  },
                  "acp_command": {
                    "type": "string",
                    "description": "Per-task ACP command override (e.g. 'claude'). Overrides the top-level acp_command for this task only."
                  },
                  "acp_args": {
                    "type": "array",
                    "items": {
                      "type": "string"
                    },
                    "description": "Per-task ACP args override."
                  },
                  "role": {
                    "type": "string",
                    "enum": [
                      "leaf",
                      "orchestrator"
                    ],
                    "description": "Per-task role override. See top-level 'role' for semantics."
                  }
                },
                "required": [
                  "goal"
                ]
              },
              "description": "Batch mode: tasks to run in parallel (limit configurable via delegation.max_concurrent_children, default 3). Each gets its own subagent with isolated context and terminal session. When provided, top-level goal/context/toolsets are ignored."
            },
            "role": {
              "type": "string",
              "enum": [
                "leaf",
                "orchestrator"
              ],
              "description": "Role of the child agent. 'leaf' (default) = focused worker, cannot delegate further. 'orchestrator' = can use delegate_task to spawn its own workers. Requires delegation.max_spawn_depth >= 2 in config; ignored (treated as 'leaf') when the child would exceed max_spawn_depth or when delegation.orchestrator_enabled=false."
            },
            "acp_command": {
              "type": "string",
              "description": "Override ACP command for child agents (e.g. 'claude', 'copilot'). When set, children use ACP subprocess transport instead of inheriting the parent's transport. Enables spawning Claude Code (claude --acp --stdio) or other ACP-capable agents from any parent, including Discord/Telegram/CLI."
            },
            "acp_args": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "description": "Arguments for the ACP command (default: ['--acp', '--stdio']). Only used when acp_command is set. Example: ['--acp', '--stdio', '--model', 'claude-opus-4-6']"
            }
          }
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "execute_code",
        "description": "Run a Python script that can call Hermes tools programmatically. Use this when you need 3+ tool calls with processing logic between them, need to filter/reduce large tool outputs before they enter your context, need conditional branching (if X then Y else Z), or need to loop (fetch N pages, process N files, retry on failure).\n\nUse normal tool calls instead when: single tool call with no processing, you need to see the full result and apply complex reasoning, or the task requires interactive user input.\n\nAvailable via `from hermes_tools import ...`:\n\n  read_file(path: str, offset: int = 1, limit: int = 500) -> dict\n    Lines are 1-indexed. Returns {\"content\": \"...\", \"total_lines\": N}\n  write_file(path: str, content: str) -> dict\n    Always overwrites the entire file.\n  search_files(pattern: str, target=\"content\", path=\".\", file_glob=None, limit=50) -> dict\n    target: \"content\" (search inside files) or \"files\" (find files by name). Returns {\"matches\": [...]}\n  patch(path: str, old_string: str, new_string: str, replace_all: bool = False) -> dict\n    Replaces old_string with new_string in the file.\n  terminal(command: str, timeout=None, workdir=None) -> dict\n    Foreground only (no background/pty). Returns {\"output\": \"...\", \"exit_code\": N}\n\nLimits: 5-minute timeout, 50KB stdout cap, max 50 tool calls per script. terminal() is foreground-only (no background or pty).\n\nScripts run in the session's working directory with the active venv's python, so project deps (pandas, etc.) and relative paths work like in terminal().\n\nPrint your final result to stdout. Use Python stdlib (json, re, math, csv, datetime, collections, etc.) for processing between tool calls.\n\nAlso available (no import needed — built into hermes_tools):\n  json_parse(text: str) — json.loads with strict=False; use for terminal() output with control chars\n  shell_quote(s: str) — shlex.quote(); use when interpolating dynamic strings into shell commands\n  retry(fn, max_attempts=3, delay=2) — retry with exponential backoff for transient failures",
        "parameters": {
          "type": "object",
          "properties": {
            "code": {
              "type": "string",
              "description": "Python code to execute. Import tools with `from hermes_tools import terminal, ...` and print your final result to stdout."
            }
          },
          "required": [
            "code"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "feishu_doc_read",
        "description": "Read the full content of a Feishu/Lark document as plain text. Useful when you need more context beyond the quoted text in a comment.",
        "parameters": {
          "type": "object",
          "properties": {
            "doc_token": {
              "type": "string",
              "description": "The document token (from the document URL or comment context)."
            }
          },
          "required": [
            "doc_token"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "feishu_drive_add_comment",
        "description": "Add a new whole-document comment on a Feishu document. Use this for whole-document comments or as a fallback when reply_comment fails with code 1069302.",
        "parameters": {
          "type": "object",
          "properties": {
            "file_token": {
              "type": "string",
              "description": "The document file token."
            },
            "content": {
              "type": "string",
              "description": "The comment text content (plain text only, no markdown)."
            },
            "file_type": {
              "type": "string",
              "description": "File type (default: docx).",
              "default": "docx"
            }
          },
          "required": [
            "file_token",
            "content"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "feishu_drive_list_comment_replies",
        "description": "List all replies in a comment thread on a Feishu document.",
        "parameters": {
          "type": "object",
          "properties": {
            "file_token": {
              "type": "string",
              "description": "The document file token."
            },
            "comment_id": {
              "type": "string",
              "description": "The comment ID to list replies for."
            },
            "file_type": {
              "type": "string",
              "description": "File type (default: docx).",
              "default": "docx"
            },
            "page_size": {
              "type": "integer",
              "description": "Number of replies per page (max 100).",
              "default": 100
            },
            "page_token": {
              "type": "string",
              "description": "Pagination token for next page."
            }
          },
          "required": [
            "file_token",
            "comment_id"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "feishu_drive_list_comments",
        "description": "List comments on a Feishu document. Use is_whole=true to list whole-document comments only.",
        "parameters": {
          "type": "object",
          "properties": {
            "file_token": {
              "type": "string",
              "description": "The document file token."
            },
            "file_type": {
              "type": "string",
              "description": "File type (default: docx).",
              "default": "docx"
            },
            "is_whole": {
              "type": "boolean",
              "description": "If true, only return whole-document comments.",
              "default": false
            },
            "page_size": {
              "type": "integer",
              "description": "Number of comments per page (max 100).",
              "default": 100
            },
            "page_token": {
              "type": "string",
              "description": "Pagination token for next page."
            }
          },
          "required": [
            "file_token"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "feishu_drive_reply_comment",
        "description": "Reply to a local comment thread on a Feishu document. Use this for local (quoted-text) comments. For whole-document comments, use feishu_drive_add_comment instead.",
        "parameters": {
          "type": "object",
          "properties": {
            "file_token": {
              "type": "string",
              "description": "The document file token."
            },
            "comment_id": {
              "type": "string",
              "description": "The comment ID to reply to."
            },
            "content": {
              "type": "string",
              "description": "The reply text content (plain text only, no markdown)."
            },
            "file_type": {
              "type": "string",
              "description": "File type (default: docx).",
              "default": "docx"
            }
          },
          "required": [
            "file_token",
            "comment_id",
            "content"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "memory",
        "description": "Save durable information to persistent memory that survives across sessions. Memory is injected into future turns, so keep it compact and focused on facts that will still matter later.\n\nWHEN TO SAVE (do this proactively, don't wait to be asked):\n- User corrects you or says 'remember this' / 'don't do that again'\n- User shares a preference, habit, or personal detail (name, role, timezone, coding style)\n- You discover something about the environment (OS, installed tools, project structure)\n- You learn a convention, API quirk, or workflow specific to this user's setup\n- You identify a stable fact that will be useful again in future sessions\n\nPRIORITY: User preferences and corrections > environment facts > procedural knowledge. The most valuable memory prevents the user from having to repeat themselves.\n\nDo NOT save task progress, session outcomes, completed-work logs, or temporary TODO state to memory; use session_search to recall those from past transcripts.\nIf you've discovered a new way to do something, solved a problem that could be necessary later, save it as a skill with the skill tool.\n\nTWO TARGETS:\n- 'user': who the user is -- name, role, preferences, communication style, pet peeves\n- 'memory': your notes -- environment facts, project conventions, tool quirks, lessons learned\n\nACTIONS: add (new entry), replace (update existing -- old_text identifies it), remove (delete -- old_text identifies it).\n\nSKIP: trivial/obvious info, things easily re-discovered, raw data dumps, and temporary task state.",
        "parameters": {
          "type": "object",
          "properties": {
            "action": {
              "type": "string",
              "enum": [
                "add",
                "replace",
                "remove"
              ],
              "description": "The action to perform."
            },
            "target": {
              "type": "string",
              "enum": [
                "memory",
                "user"
              ],
              "description": "Which memory store: 'memory' for personal notes, 'user' for user profile."
            },
            "content": {
              "type": "string",
              "description": "The entry content. Required for 'add' and 'replace'."
            },
            "old_text": {
              "type": "string",
              "description": "Short unique substring identifying the entry to replace or remove."
            }
          },
          "required": [
            "action",
            "target"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "patch",
        "description": "Targeted find-and-replace edits in files. Use this instead of sed/awk in terminal. Uses fuzzy matching (9 strategies) so minor whitespace/indentation differences won't break it. Returns a unified diff. Auto-runs syntax checks after editing.\n\nReplace mode (default): find a unique string and replace it.\nPatch mode: apply V4A multi-file patches for bulk changes.",
        "parameters": {
          "type": "object",
          "properties": {
            "mode": {
              "type": "string",
              "enum": [
                "replace",
                "patch"
              ],
              "description": "Edit mode: 'replace' for targeted find-and-replace, 'patch' for V4A multi-file patches",
              "default": "replace"
            },
            "path": {
              "type": "string",
              "description": "File path to edit (required for 'replace' mode)"
            },
            "old_string": {
              "type": "string",
              "description": "Text to find in the file (required for 'replace' mode). Must be unique in the file unless replace_all=true. Include enough surrounding context to ensure uniqueness."
            },
            "new_string": {
              "type": "string",
              "description": "Replacement text (required for 'replace' mode). Can be empty string to delete the matched text."
            },
            "replace_all": {
              "type": "boolean",
              "description": "Replace all occurrences instead of requiring a unique match (default: false)",
              "default": false
            },
            "patch": {
              "type": "string",
              "description": "V4A format patch content (required for 'patch' mode). Format:\n*** Begin Patch\n*** Update File: path/to/file\n@@ context hint @@\n context line\n-removed line\n+added line\n*** End Patch"
            }
          },
          "required": [
            "mode"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "process",
        "description": "Manage background processes started with terminal(background=true). Actions: 'list' (show all), 'poll' (check status + new output), 'log' (full output with pagination), 'wait' (block until done or timeout), 'kill' (terminate), 'write' (send raw stdin data without newline), 'submit' (send data + Enter, for answering prompts), 'close' (close stdin/send EOF).",
        "parameters": {
          "type": "object",
          "properties": {
            "action": {
              "type": "string",
              "enum": [
                "list",
                "poll",
                "log",
                "wait",
                "kill",
                "write",
                "submit",
                "close"
              ],
              "description": "Action to perform on background processes"
            },
            "session_id": {
              "type": "string",
              "description": "Process session ID (from terminal background output). Required for all actions except 'list'."
            },
            "data": {
              "type": "string",
              "description": "Text to send to process stdin (for 'write' and 'submit' actions)"
            },
            "timeout": {
              "type": "integer",
              "description": "Max seconds to block for 'wait' action. Returns partial output on timeout.",
              "minimum": 1
            },
            "offset": {
              "type": "integer",
              "description": "Line offset for 'log' action (default: last 200 lines)"
            },
            "limit": {
              "type": "integer",
              "description": "Max lines to return for 'log' action",
              "minimum": 1
            }
          },
          "required": [
            "action"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "read_file",
        "description": "Read a text file with line numbers and pagination. Use this instead of cat/head/tail in terminal. Output format: 'LINE_NUM|CONTENT'. Suggests similar filenames if not found. Use offset and limit for large files. Reads exceeding ~100K characters are rejected; use offset and limit to read specific sections of large files. NOTE: Cannot read images or binary files — use vision_analyze for images.",
        "parameters": {
          "type": "object",
          "properties": {
            "path": {
              "type": "string",
              "description": "Path to the file to read (absolute, relative, or ~/path)"
            },
            "offset": {
              "type": "integer",
              "description": "Line number to start reading from (1-indexed, default: 1)",
              "default": 1,
              "minimum": 1
            },
            "limit": {
              "type": "integer",
              "description": "Maximum number of lines to read (default: 500, max: 2000)",
              "default": 500,
              "maximum": 2000
            }
          },
          "required": [
            "path"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "search_files",
        "description": "Search file contents or find files by name. Use this instead of grep/rg/find/ls in terminal. Ripgrep-backed, faster than shell equivalents.\n\nContent search (target='content'): Regex search inside files. Output modes: full matches with line numbers, file paths only, or match counts.\n\nFile search (target='files'): Find files by glob pattern (e.g., '*.py', '*config*'). Also use this instead of ls — results sorted by modification time.",
        "parameters": {
          "type": "object",
          "properties": {
            "pattern": {
              "type": "string",
              "description": "Regex pattern for content search, or glob pattern (e.g., '*.py') for file search"
            },
            "target": {
              "type": "string",
              "enum": [
                "content",
                "files"
              ],
              "description": "'content' searches inside file contents, 'files' searches for files by name",
              "default": "content"
            },
            "path": {
              "type": "string",
              "description": "Directory or file to search in (default: current working directory)",
              "default": "."
            },
            "file_glob": {
              "type": "string",
              "description": "Filter files by pattern in grep mode (e.g., '*.py' to only search Python files)"
            },
            "limit": {
              "type": "integer",
              "description": "Maximum number of results to return (default: 50)",
              "default": 50
            },
            "offset": {
              "type": "integer",
              "description": "Skip first N results for pagination (default: 0)",
              "default": 0
            },
            "output_mode": {
              "type": "string",
              "enum": [
                "content",
                "files_only",
                "count"
              ],
              "description": "Output format for grep mode: 'content' shows matching lines with line numbers, 'files_only' lists file paths, 'count' shows match counts per file",
              "default": "content"
            },
            "context": {
              "type": "integer",
              "description": "Number of context lines before and after each match (grep mode only)",
              "default": 0
            }
          },
          "required": [
            "pattern"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "send_message",
        "description": "Send a message to a connected messaging platform, or list available targets.\n\nIMPORTANT: When the user asks to send to a specific channel or person (not just a bare platform name), call send_message(action='list') FIRST to see available targets, then send to the correct one.\nIf the user just says a platform name like 'send to telegram', send directly to the home channel without listing first.",
        "parameters": {
          "type": "object",
          "properties": {
            "action": {
              "type": "string",
              "enum": [
                "send",
                "list"
              ],
              "description": "Action to perform. 'send' (default) sends a message. 'list' returns all available channels/contacts across connected platforms."
            },
            "target": {
              "type": "string",
              "description": "Delivery target. Format: 'platform' (uses home channel), 'platform:#channel-name', 'platform:chat_id', or 'platform:chat_id:thread_id' for Telegram topics and Discord threads. Examples: 'telegram', 'telegram:-1001234567890:17585', 'discord:999888777:555444333', 'discord:#bot-home', 'slack:#engineering', 'signal:+155****4567', 'matrix:!roomid:server.org', 'matrix:@user:server.org', 'yuanbao:direct:<account_id>' (DM), 'yuanbao:group:<group_code>' (group chat)"
            },
            "message": {
              "type": "string",
              "description": "The message text to send. To send an image or file, include MEDIA:<local_path> (e.g. 'MEDIA:/tmp/hermes/cache/img_xxx.jpg') in the message — the platform will deliver it as a native media attachment."
            }
          }
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "session_search",
        "description": "Search your long-term memory of past conversations, or browse recent sessions. This is your recall -- every past session is searchable, and this tool summarizes what happened.\n\nTWO MODES:\n1. Recent sessions (no query): Call with no arguments to see what was worked on recently. Returns titles, previews, and timestamps. Zero LLM cost, instant. Start here when the user asks what were we working on or what did we do recently.\n2. Keyword search (with query): Search for specific topics across all past sessions. Returns LLM-generated summaries of matching sessions.\n\nUSE THIS PROACTIVELY when:\n- The user says 'we did this before', 'remember when', 'last time', 'as I mentioned'\n- The user asks about a topic you worked on before but don't have in current context\n- The user references a project, person, or concept that seems familiar but isn't in memory\n- You want to check if you've solved a similar problem before\n- The user asks 'what did we do about X?' or 'how did we fix Y?'\n\nDon't hesitate to search when it is actually cross-session -- it's fast and cheap. Better to search and confirm than to guess or ask the user to repeat themselves.\n\nSearch syntax: keywords joined with OR for broad recall (elevenlabs OR baseten OR funding), phrases for exact match (\"docker networking\"), boolean (python NOT java), prefix (deploy*). IMPORTANT: Use OR between keywords for best results — FTS5 defaults to AND which misses sessions that only mention some terms. If a broad OR query returns nothing, try individual keyword searches in parallel. Returns summaries of the top matching sessions.",
        "parameters": {
          "type": "object",
          "properties": {
            "query": {
              "type": "string",
              "description": "Search query — keywords, phrases, or boolean expressions to find in past sessions. Omit this parameter entirely to browse recent sessions instead (returns titles, previews, timestamps with no LLM cost)."
            },
            "role_filter": {
              "type": "string",
              "description": "Optional: only search messages from specific roles (comma-separated). E.g. 'user,assistant' to skip tool outputs."
            },
            "limit": {
              "type": "integer",
              "description": "Max sessions to summarize (default: 3, max: 5).",
              "default": 3
            }
          }
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "skill_manage",
        "description": "Manage skills (create, update, delete). Skills are your procedural memory — reusable approaches for recurring task types. New skills go to ~/.hermes/skills/; existing skills can be modified wherever they live.\n\nActions: create (full SKILL.md + optional category), patch (old_string/new_string — preferred for fixes), edit (full SKILL.md rewrite — major overhauls only), delete, write_file, remove_file.\n\nCreate when: complex task succeeded (5+ calls), errors overcome, user-corrected approach worked, non-trivial workflow discovered, or user asks you to remember a procedure.\nUpdate when: instructions stale/wrong, OS-specific failures, missing steps or pitfalls found during use. If you used a skill and hit issues not covered by it, patch it immediately.\n\nAfter difficult/iterative tasks, offer to save as a skill. Skip for simple one-offs. Confirm with user before creating/deleting.\n\nGood skills: trigger conditions, numbered steps with exact commands, pitfalls section, verification steps. Use skill_view() to see format examples.\n\nPinned skills are off-limits — all write actions refuse with a message pointing the user to `hermes curator unpin <name>`. Don't try to route around this by renaming or recreating.",
        "parameters": {
          "type": "object",
          "properties": {
            "action": {
              "type": "string",
              "enum": [
                "create",
                "patch",
                "edit",
                "delete",
                "write_file",
                "remove_file"
              ],
              "description": "The action to perform."
            },
            "name": {
              "type": "string",
              "description": "Skill name (lowercase, hyphens/underscores, max 64 chars). Must match an existing skill for patch/edit/delete/write_file/remove_file."
            },
            "content": {
              "type": "string",
              "description": "Full SKILL.md content (YAML frontmatter + markdown body). Required for 'create' and 'edit'. For 'edit', read the skill first with skill_view() and provide the complete updated text."
            },
            "old_string": {
              "type": "string",
              "description": "Text to find in the file (required for 'patch'). Must be unique unless replace_all=true. Include enough surrounding context to ensure uniqueness."
            },
            "new_string": {
              "type": "string",
              "description": "Replacement text (required for 'patch'). Can be empty string to delete the matched text."
            },
            "replace_all": {
              "type": "boolean",
              "description": "For 'patch': replace all occurrences instead of requiring a unique match (default: false)."
            },
            "category": {
              "type": "string",
              "description": "Optional category/domain for organizing the skill (e.g., 'devops', 'data-science', 'mlops'). Creates a subdirectory grouping. Only used with 'create'."
            },
            "file_path": {
              "type": "string",
              "description": "Path to a supporting file within the skill directory. For 'write_file'/'remove_file': required, must be under references/, templates/, scripts/, or assets/. For 'patch': optional, defaults to SKILL.md if omitted."
            },
            "file_content": {
              "type": "string",
              "description": "Content for the file. Required for 'write_file'."
            }
          },
          "required": [
            "action",
            "name"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "skill_view",
        "description": "Skills allow for loading information about specific tasks and workflows, as well as scripts and templates. Load a skill's full content or access its linked files (references, templates, scripts). First call returns SKILL.md content plus a 'linked_files' dict showing available references/templates/scripts. To access those, call again with file_path parameter.",
        "parameters": {
          "type": "object",
          "properties": {
            "name": {
              "type": "string",
              "description": "The skill name (use skills_list to see available skills). For plugin-provided skills, use the qualified form 'plugin:skill' (e.g. 'superpowers:writing-plans')."
            },
            "file_path": {
              "type": "string",
              "description": "OPTIONAL: Path to a linked file within the skill (e.g., 'references/api.md', 'templates/config.yaml', 'scripts/validate.py'). Omit to get the main SKILL.md content."
            }
          },
          "required": [
            "name"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "skills_list",
        "description": "List available skills (name + description). Use skill_view(name) to load full content.",
        "parameters": {
          "type": "object",
          "properties": {
            "category": {
              "type": "string",
              "description": "Optional category filter to narrow results"
            }
          }
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "terminal",
        "description": "Execute shell commands on a Linux environment. Filesystem usually persists between calls.\n\nDo NOT use cat/head/tail to read files — use read_file instead.\nDo NOT use grep/rg/find to search — use search_files instead.\nDo NOT use ls to list directories — use search_files(target='files') instead.\nDo NOT use sed/awk to edit files — use patch instead.\nDo NOT use echo/cat heredoc to create files — use write_file instead.\nReserve terminal for: builds, installs, git, processes, scripts, network, package managers, and anything that needs a shell.\n\nForeground (default): Commands return INSTANTLY when done, even if the timeout is high. Set timeout=300 for long builds/scripts — you'll still get the result in seconds if it's fast. Prefer foreground for short commands.\nBackground: Set background=true to get a session_id. Two patterns:\n  (1) Long-lived processes that never exit (servers, watchers).\n  (2) Long-running tasks with notify_on_complete=true — you can keep working on other things and the system auto-notifies you when the task finishes. Great for test suites, builds, deployments, or anything that takes more than a minute.\nFor servers/watchers, do NOT use shell-level background wrappers (nohup/disown/setsid/trailing '&') in foreground mode. Use background=true so Hermes can track lifecycle and output.\nAfter starting a server, verify readiness with a health check or log signal, then run tests in a separate terminal() call. Avoid blind sleep loops.\nUse process(action=\"poll\") for progress checks, process(action=\"wait\") to block until done.\nWorking directory: Use 'workdir' for per-command cwd.\nPTY mode: Set pty=true for interactive CLI tools (Codex, Claude Code, Python REPL).\n\nDo NOT use vim/nano/interactive tools without pty=true — they hang without a pseudo-terminal. Pipe git output to cat if it might page.\n",
        "parameters": {
          "type": "object",
          "properties": {
            "command": {
              "type": "string",
              "description": "The command to execute on the VM"
            },
            "background": {
              "type": "boolean",
              "description": "Run the command in the background. Two patterns: (1) Long-lived processes that never exit (servers, watchers). (2) Long-running tasks paired with notify_on_complete=true — you can keep working and get notified when the task finishes. For short commands, prefer foreground with a generous timeout instead.",
              "default": false
            },
            "timeout": {
              "type": "integer",
              "description": "Max seconds to wait (default: 180, foreground max: 600). Returns INSTANTLY when command finishes — set high for long tasks, you won't wait unnecessarily. Foreground timeout above 600s is rejected; use background=true for longer commands.",
              "minimum": 1
            },
            "workdir": {
              "type": "string",
              "description": "Working directory for this command (absolute path). Defaults to the session working directory."
            },
            "pty": {
              "type": "boolean",
              "description": "Run in pseudo-terminal (PTY) mode for interactive CLI tools like Codex, Claude Code, or Python REPL. Only works with local and SSH backends. Default: false.",
              "default": false
            },
            "notify_on_complete": {
              "type": "boolean",
              "description": "When true (and background=true), you'll be automatically notified exactly once when the process finishes. **This is the right choice for almost every long-running task** — tests, builds, deployments, multi-item batch jobs, anything that takes over a minute and has a defined end. Use this and keep working on other things; the system notifies you on exit. MUTUALLY EXCLUSIVE with watch_patterns — when both are set, watch_patterns is dropped.",
              "default": false
            },
            "watch_patterns": {
              "type": "array",
              "items": {
                "type": "string"
              },
              "description": "Strings to watch for in background process output. HARD RATE LIMIT: at most 1 notification per 15 seconds per process — matches arriving inside the cooldown are dropped. After 3 consecutive 15-second windows with dropped matches, watch_patterns is automatically disabled for that process and promoted to notify_on_complete behavior (one notification on exit, no more mid-process spam). USE ONLY for truly rare, one-shot mid-process signals on LONG-LIVED processes that will never exit on their own — e.g. ['Application startup complete'] on a server so you know when to hit its endpoint, or ['migration done'] on a daemon. DO NOT use for: (1) end-of-run markers like 'DONE'/'PASS' — use notify_on_complete instead; (2) error patterns like 'ERROR'/'Traceback' in loops or multi-item batch jobs — they fire on every iteration and you'll hit the strike limit fast; (3) anything you'd ever combine with notify_on_complete. When in doubt, choose notify_on_complete. MUTUALLY EXCLUSIVE with notify_on_complete — set one, not both."
            }
          },
          "required": [
            "command"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "text_to_speech",
        "description": "Convert text to speech audio. Returns a MEDIA: path that the platform delivers as native audio. Compatible providers render as a voice bubble on Telegram; otherwise audio is sent as a regular attachment. In CLI mode, saves to ~/voice-memos/. Voice and provider are user-configured (built-in providers like edge/openai or custom command providers under tts.providers.<name>), not model-selected.",
        "parameters": {
          "type": "object",
          "properties": {
            "text": {
              "type": "string",
              "description": "The text to convert to speech. Provider-specific character caps apply and are enforced automatically (OpenAI 4096, xAI 15000, MiniMax 10000, ElevenLabs 5k-40k depending on model); over-long input is truncated."
            },
            "output_path": {
              "type": "string",
              "description": "Optional custom file path to save the audio. Defaults to ~/.hermes/audio_cache/<timestamp>.mp3"
            }
          },
          "required": [
            "text"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "todo",
        "description": "Manage your task list for the current session. Use for complex tasks with 3+ steps or when the user provides multiple tasks. Call with no parameters to read the current list.\n\nWriting:\n- Provide 'todos' array to create/update items\n- merge=false (default): replace the entire list with a fresh plan\n- merge=true: update existing items by id, add any new ones\n\nEach item: {id: string, content: string, status: pending|in_progress|completed|cancelled}\nList order is priority. Only ONE item in_progress at a time.\nMark items completed immediately when done. If something fails, cancel it and add a revised item.\n\nAlways returns the full current list.",
        "parameters": {
          "type": "object",
          "properties": {
            "todos": {
              "type": "array",
              "description": "Task items to write. Omit to read current list.",
              "items": {
                "type": "object",
                "properties": {
                  "id": {
                    "type": "string",
                    "description": "Unique item identifier"
                  },
                  "content": {
                    "type": "string",
                    "description": "Task description"
                  },
                  "status": {
                    "type": "string",
                    "enum": [
                      "pending",
                      "in_progress",
                      "completed",
                      "cancelled"
                    ],
                    "description": "Current status"
                  }
                },
                "required": [
                  "id",
                  "content",
                  "status"
                ]
              }
            },
            "merge": {
              "type": "boolean",
              "description": "true: update existing items by id, add new ones. false (default): replace the entire list.",
              "default": false
            }
          }
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "vision_analyze",
        "description": "Inspect an image from a URL, file path, or tool output when you need closer detail than what's visible in the conversation. If the user's image is already attached to the conversation and you can see it, just answer directly — only call this tool for images referenced by URL/path, images returned inside other tool results (browser screenshots, search thumbnails), or when you need a deeper look at a specific region the main model's vision may have missed.",
        "parameters": {
          "type": "object",
          "properties": {
            "image_url": {
              "type": "string",
              "description": "Image URL (http/https) or local file path to analyze."
            },
            "question": {
              "type": "string",
              "description": "Your specific question or request about the image to resolve. The AI will automatically provide a complete image description AND answer your specific question."
            }
          },
          "required": [
            "image_url",
            "question"
          ]
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "write_file",
        "description": "Write content to a file, completely replacing existing content. Use this instead of echo/cat heredoc in terminal. Creates parent directories automatically. OVERWRITES the entire file — use 'patch' for targeted edits.",
        "parameters": {
          "type": "object",
          "properties": {
            "path": {
              "type": "string",
              "description": "Path to the file to write (will be created if it doesn't exist, overwritten if it does)"
            },
            "content": {
              "type": "string",
              "description": "Complete content to write to the file"
            }
          },
          "required": [
            "path",
            "content"
          ]
        }
      }
    }
  ],
  "message_count": 6,
  "messages": [
    {
      "role": "user",
      "content": "You are running as Hermes' background skill CURATOR. This is an UMBRELLA-BUILDING consolidation pass, not a passive audit and not a duplicate-finder.\n\nThe goal of the skill collection is a LIBRARY OF CLASS-LEVEL INSTRUCTIONS AND EXPERIENTIAL KNOWLEDGE. A collection of hundreds of narrow skills where each one captures one session's specific bug is a FAILURE of the library — not a feature. An agent searching skills matches on descriptions, not on exact names; one broad umbrella skill with labeled subsections beats five narrow siblings for discoverability, not the other way around.\n\nThe right target shape is CLASS-LEVEL skills with rich SKILL.md bodies + `references/`, `templates/`, and `scripts/` subfiles for session-specific detail — not one-session-one-skill micro-entries.\n\nHard rules — do not violate:\n1. DO NOT touch bundled or hub-installed skills. The candidate list below is already filtered to agent-created skills only.\n2. DO NOT delete any skill. Archiving (moving the skill's directory into ~/.hermes/skills/.archive/) is the maximum destructive action. Archives are recoverable; deletion is not.\n3. DO NOT touch skills shown as pinned=yes. Skip them entirely.\n4. DO NOT use usage counters as a reason to skip consolidation. The counters are new and often mostly zero. Judge overlap on CONTENT, not on use_count. 'use=0' is not evidence a skill is valuable; it's absence of evidence either way.\n5. DO NOT reject consolidation on the grounds that 'each skill has a distinct trigger'. Pairwise distinctness is the wrong bar. The right bar is: 'would a human maintainer write this as N separate skills, or as one skill with N labeled subsections?' When the answer is the latter, merge.\n\nHow to work — not optional:\n1. Scan the full candidate list. Identify PREFIX CLUSTERS (skills sharing a first word or domain keyword). Examples you are likely to find: hermes-config-*, hermes-dashboard-*, gateway-*, codex-*, ollama-*, anthropic-*, gemini-*, mcp-*, salvage-*, pr-*, competitor-*, python-*, security-*, etc. Expect 10-25 clusters.\n2. For each cluster with 2+ members, do NOT ask 'are these pairs overlapping?' — ask 'what is the UMBRELLA CLASS these skills all serve? Would a maintainer name that class and write one skill for it?' If yes, pick (or create) the umbrella and absorb the siblings into it.\n3. Three ways to consolidate — use the right one per cluster:\n   a. MERGE INTO EXISTING UMBRELLA — one skill in the cluster is already broad enough to be the umbrella (example: `pr-triage-salvage` for the PR review cluster). Patch it to add a labeled section for each sibling's unique insight, then archive the siblings.\n   b. CREATE A NEW UMBRELLA SKILL.md — no existing member is broad enough. Use skill_manage action=create to write a new class-level skill whose SKILL.md covers the shared workflow and has short labeled subsections. Archive the now-absorbed narrow siblings.\n   c. DEMOTE TO REFERENCES/TEMPLATES/SCRIPTS — a sibling has narrow-but-valuable session-specific content. Move it into the umbrella's appropriate support directory:\n      • `references/<topic>.md` for session-specific detail OR condensed knowledge banks (quoted research, API docs excerpts, domain notes, provider quirks, reproduction recipes)\n      • `templates/<name>.<ext>` for starter files meant to be copied and modified\n      • `scripts/<name>.<ext>` for statically re-runnable actions (verification scripts, fixture generators, probes)\n      Then archive the old sibling. Use `terminal` with `mkdir -p ~/.hermes/skills/<umbrella>/references/ && mv ... <umbrella>/references/<topic>.md` (or templates/ / scripts/).\n4. Also flag skills whose NAME is too narrow (contains a PR number, a feature codename, a specific error string, an 'audit' / 'diagnosis' / 'salvage' session artifact). These almost always belong as a subsection or support file under a class-level umbrella.\n5. Iterate. After one consolidation round, scan the remaining set and look for the NEXT umbrella opportunity. Don't stop after 3 merges.\n\nYour toolset:\n  - skills_list, skill_view        — read the current landscape\n  - skill_manage action=patch      — add sections to the umbrella\n  - skill_manage action=create     — create a new umbrella SKILL.md\n  - skill_manage action=write_file — add a references/, templates/, or scripts/ file under an existing skill (the skill must already exist)\n  - terminal                       — mv a sibling into the archive OR move its content into a support subfile\n\n'keep' is a legitimate decision ONLY when the skill is already a class-level umbrella and none of the proposed merges would improve discoverability. 'This is narrow but distinct from its siblings' is NOT a reason to keep — it's a reason to move it under an umbrella as a subsection or support file.\n\nExpected output: real umbrella-ification. Process every obvious cluster. If you end the pass with fewer than 10 archives, you stopped too early — go back and look at the clusters you left alone.\n\nWhen done, write a human summary AND a structured machine-readable block so downstream tooling can distinguish consolidation from pruning. Format EXACTLY:\n\n## Structured summary (required)\n```yaml\nconsolidations:\n  - from: <old-skill-name>\n    into: <umbrella-skill-name>\n    reason: <one short sentence — why merged, not just 'similar'>\nprunings:\n  - name: <skill-name>\n    reason: <one short sentence — why archived with no merge target>\n```\n\nEvery skill you moved to .archive/ MUST appear in exactly one of the two lists. If you consolidated X into umbrella Y (patched Y, wrote a references file to Y, or created Y with X's content absorbed), X goes under `consolidations` with `into: Y`. If you archived X with no absorption — truly stale, irrelevant, or obsolete — X goes under `prunings`. Leave a list empty (`consolidations: []`) if none. Do not omit the block. The block comes AFTER your human-readable summary of clusters processed, patches made, and decisions left alone.\n\nAgent-created skills (3):\n\n- docker  state=active  pinned=no  activity=2  use=1  view=1  patches=0  last_activity=2026-05-23T09:49:35.204714+00:00\n- hyperframes-deploy  state=active  pinned=no  activity=2  use=1  view=1  patches=0  last_activity=2026-05-23T09:45:39.876476+00:00\n- python-webapp-deploy  state=active  pinned=no  activity=24  use=6  view=6  patches=12  last_activity=2026-05-25T06:42:29.888115+00:00"
    },
    {
      "role": "assistant",
      "content": "I'll begin by reviewing the full candidate set. Only 3 skills — let me load each one to assess overlap and clustering potential.",
      "reasoning": "Let me start by examining the three agent-created skills to understand their content and identify any clustering opportunities.",
      "finish_reason": "tool_calls",
      "reasoning_content": "Let me start by examining the three agent-created skills to understand their content and identify any clustering opportunities.",
      "tool_calls": [
        {
          "id": "call_df95c5969a39497aa6393820",
          "call_id": "call_df95c5969a39497aa6393820",
          "response_item_id": "fc_df95c5969a39497aa6393820",
          "type": "function",
          "function": {
            "name": "skill_view",
            "arguments": "{\"name\": \"docker\"}"
          }
        },
        {
          "id": "call_cae4c7e1f34a44f28dacada3",
          "call_id": "call_cae4c7e1f34a44f28dacada3",
          "response_item_id": "fc_cae4c7e1f34a44f28dacada3",
          "type": "function",
          "function": {
            "name": "skill_view",
            "arguments": "{\"name\": \"hyperframes-deploy\"}"
          }
        },
        {
          "id": "call_e62ab854d94b4509a31d23dd",
          "call_id": "call_e62ab854d94b4509a31d23dd",
          "response_item_id": "fc_e62ab854d94b4509a31d23dd",
          "type": "function",
          "function": {
            "name": "skill_view",
            "arguments": "{\"name\": \"python-webapp-deploy\"}"
          }
        }
      ]
    },
    {
      "role": "tool",
      "content": "{\"success\": true, \"name\": \"docker\", \"description\": \"Docker on Linux: install Engine + Compose, configure proxy (China/mihomo), run databases (PostgreSQL, MySQL, KingbaseES), MinIO object storage, deploy full-stack apps. Pitfalls: sudo-via-pipe, GPG keys, socket permissions, root-can't-run-db.\", \"tags\": [\"docker\", \"devops\", \"linux\", \"containers\", \"database\", \"postgresql\", \"mysql\", \"kingbase\", \"minio\", \"proxy\", \"deployment\", \"full-stack\"], \"related_skills\": [], \"content\": \"---\\nname: docker\\ndescription: \\\"Docker on Linux: install Engine + Compose, configure proxy (China/mihomo), run databases (PostgreSQL, MySQL, KingbaseES), MinIO object storage, deploy full-stack apps. Pitfalls: sudo-via-pipe, GPG keys, socket permissions, root-can't-run-db.\\\"\\ntags: [docker, devops, linux, containers, database, postgresql, mysql, kingbase, minio, proxy, deployment, full-stack]\\ntriggers:\\n  - install docker\\n  - setup docker\\n  - docker engine\\n  - docker compose\\n  - database docker container\\n  - run database in docker\\n  - restore database docker\\n  - kingbase docker\\n  - postgresql docker\\n  - mysql docker container\\n  - minio docker\\n  - object storage docker\\n  - full stack deploy docker\\n  - prisma docker\\n  - docker proxy\\n  - docker hub unreachable\\n  - docker permission denied\\n---\\n\\n# Docker: Install, Configure, and Run Services\\n\\nComplete Docker workflow on Linux — from installation through running databases, object storage, and full-stack applications in containers.\\n\\n---\\n\\n## Part 1: Installation (Ubuntu/Debian)\\n\\n### Prerequisites Check\\n\\n```bash\\ncat /etc/os-release | head -5   # OS version\\nwhich docker 2>/dev/null        # already installed?\\nwhoami && id                    # current user\\n```\\n\\n### Standard Install (requires root/sudo)\\n\\n**When sudo needs a password:** Use `echo 'PASSWORD' | sudo -S COMMAND` pattern. See pitfalls below.\\n\\n1. **Install dependencies**\\n   ```bash\\n   sudo apt-get update -y\\n   sudo apt-get install -y ca-certificates curl gnupg\\n   ```\\n\\n2. **Add Docker GPG key** — download first, then dearmor (do NOT pipe curl into gpg when using sudo-via-pipe)\\n   ```bash\\n   curl -fsSL https://download.docker.com/linux/ubuntu/gpg -o /tmp/docker.gpg\\n   sudo install -m 0755 -d /etc/apt/keyrings\\n   sudo gpg --batch --yes --dearmor -o /etc/apt/keyrings/docker.gpg /tmp/docker.gpg\\n   sudo chmod a+r /etc/apt/keyrings/docker.gpg\\n   ```\\n\\n3. **Add Docker apt repo** — write to temp file, then copy (tee consumes stdin)\\n   ```bash\\n   ARCH=$(dpkg --print-architecture)\\n   echo \\\"deb [arch=$ARCH signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu $(lsb_release -cs) stable\\\" > /tmp/docker.list\\n   sudo cp /tmp/docker.list /etc/apt/sources.list.d/docker.list\\n   ```\\n\\n4. **Install Docker packages**\\n   ```bash\\n   sudo apt-get update -y\\n   sudo apt-get install -y docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin\\n   ```\\n\\n5. **Post-install setup**\\n   ```bash\\n   sudo usermod -aG docker $USER    # add user to docker group\\n   sudo systemctl start docker\\n   sudo systemctl enable docker\\n   ```\\n\\n6. **Verify**\\n   ```bash\\n   docker --version\\n   docker compose version\\n   docker info | head -5\\n   ```\\n\\n### Rootless Docker (no root needed)\\n\\nFor environments without sudo access. Requires `/etc/subuid` and `/etc/subgid` entries for the user.\\n\\n```bash\\ncat /etc/subuid | grep $(whoami)\\ncat /etc/subgid | grep $(whoami)\\n# If missing, ask admin to add: username:100000:65536\\n\\ndockerd-rootless-setuptool.sh install\\n```\\n\\n### Passwordless sudo (convenience)\\n\\n```bash\\nsudo usermod -aG sudo $USER\\nsudo bash -c 'echo \\\"$USER ALL=(ALL) NOPASSWD: ALL\\\" > /etc/sudoers.d/$USER && chmod 440 /etc/sudoers.d/$USER'\\n```\\n\\n⚠️ Full NOPASSWD sudo = unrestricted root. Fine for dev machines, risky in production.\\n\\n---\\n\\n## Part 2: Proxy Configuration (China / Restricted Networks)\\n\\nChina Docker mirrors are unreliable (rate-limited, frequently down). Using a local proxy (mihomo/Clash) is more reliable.\\n\\n### Find proxy port\\n\\n```bash\\nps aux | grep -i clash | grep -v grep\\nss -tlnp | grep -E '7890|7891|7892'\\n```\\n\\nCommon mihomo/Clash ports:\\n- `7890` — mixed (HTTP + SOCKS5) ← **use this one**\\n- `7891` — HTTP only\\n- `7892` — SOCKS5 only\\n- `9090` — API/dashboard\\n\\n### Verify proxy works for Docker Hub\\n\\n```bash\\ncurl -x http://127.0.0.1:7890 -s --connect-timeout 10 https://registry-1.docker.io/v2/\\n# Should return: {\\\"errors\\\":[{\\\"code\\\":\\\"UNAUTHORIZED\\\",...}]}  ← means it works\\n```\\n\\n### Configure Docker systemd proxy\\n\\n```bash\\nsudo mkdir -p /etc/systemd/system/docker.service.d\\nsudo tee /etc/systemd/system/docker.service.d/proxy.conf <<'EOF'\\n[Service]\\nEnvironment=\\\"HTTP_PROXY=http://127.0.0.1:7890\\\"\\nEnvironment=\\\"HTTPS_PROXY=http://127.0.0.1:7890\\\"\\nEnvironment=\\\"NO_PROXY=localhost,127.0.0.1,10.0.0.0/8,172.16.0.0/12,192.168.0.0/16\\\"\\nEOF\\nsudo systemctl daemon-reload && sudo systemctl restart docker\\n```\\n\\n> **Note:** Do NOT put proxy in `/etc/docker/daemon.json` — that file is for registry mirrors and other daemon config. Systemd override is the correct place for proxy env vars.\\n\\n### Clear mirror config if present\\n\\n```bash\\nsudo tee /etc/docker/daemon.json <<'EOF'\\n{}\\nEOF\\nsudo systemctl restart docker\\n```\\n\\n### Test proxy connectivity\\n\\n```bash\\ncurl -x http://127.0.0.1:7890 -s --connect-timeout 5 https://registry-1.docker.io/v2/\\n```\\n\\n**Clash/mihomo:** Use the mixed port (7890) rather than the HTTP port (7891) — the mixed port handles both HTTP and HTTPS better.\\n\\n### China Mirror Fallback (if no proxy available)\\n\\n```json\\n// /etc/docker/daemon.json\\n{\\n  \\\"registry-mirrors\\\": [\\n    \\\"https://mirror.ccs.tencentyun.com\\\",\\n    \\\"https://registry.docker-cn.com\\\",\\n    \\\"https://docker.1ms.run\\\"\\n  ]\\n}\\n```\\nExpect rate limiting (429) on many mirrors.\\n\\n---\\n\\n## Part 3: Running Database Containers\\n\\n### General Workflow\\n\\n#### 1. Pull and inspect the image\\n\\n```bash\\ndocker pull IMAGE:TAG\\ndocker inspect IMAGE:TAG --format '{{json .Config.Entrypoint}}'\\ndocker inspect IMAGE:TAG --format '{{json .Config.Env}}'\\n```\\n\\nAlways inspect the entrypoint script and env vars — different images use different variable names for user/password/database.\\n\\n#### 2. Start the container\\n\\n```bash\\ndocker run -d \\\\\\n  --name db-container \\\\\\n  --user <db_user> \\\\          # run as the database's expected user, NOT root\\n  -p 54321:54321 \\\\\\n  -e DB_USER=myuser \\\\\\n  -e DB_PASSWORD=mypass \\\\\\n  -v /host/data:/container/data \\\\\\n  IMAGE:TAG\\n```\\n\\n#### 3. Verify startup\\n\\n```bash\\ndocker ps | grep db-container\\ndocker logs db-container 2>&1 | tail -20\\n```\\n\\n#### 4. Create databases and users\\n\\n```bash\\ndocker exec db-container bash -c 'source /etc/profile && TOOL -U admin_user -d default_db -c \\\"CREATE DATABASE mydb;\\\"'\\n```\\n\\n### Critical: User Permission Issues\\n\\n**Root cannot run database engines.** Most databases (PostgreSQL, KingbaseES, MySQL) **refuse to run as root** for security:\\n```\\ninitdb: error: cannot be run as root\\n```\\n\\n**Do NOT use `--user root`**. Use the image's intended user (often `postgres`, `kingbase`, `mysql`).\\n\\n**`--privileged` flag:** Useful when running as the image's intended user (e.g., `kingbase`) and the entrypoint script uses `sudo` heavily. Helps avoid PAM permission errors without needing root:\\n```bash\\ndocker run -d --name kingbasev8 --user kingbase --privileged -p 54321:54321 ...\\n```\\nDo NOT combine `--privileged` with `--user root` — same initdb issue.\\n\\n### sudo PAM errors in container logs\\n\\nIf the entrypoint script uses `sudo` and you see:\\n```\\nsudo: pam_open_session: Permission denied\\nsudo: policy plugin failed session initialization\\n```\\n\\nThis is usually **non-critical** — it affects cron/logrotate setup, not the database itself. Check if the database actually started:\\n```\\nwaiting for server to start.... done\\nserver started   ← this means it's fine\\n```\\n\\n### Environment Variable Discovery\\n\\nDifferent images use different env var names. **Always inspect the entrypoint script**:\\n\\n```bash\\ndocker exec db-container cat /home/USER/docker-entrypoint.sh\\n```\\n\\n| Image | User var | Password var | Database var |\\n|-------|----------|-------------|-------------|\\n| PostgreSQL | `POSTGRES_USER` | `POSTGRES_PASSWORD` | `POSTGRES_DB` |\\n| KingbaseES | `DB_USER` | `DB_PASSWORD` or `PASSWORD` | (created manually) |\\n| MySQL | `MYSQL_USER` | `MYSQL_PASSWORD` | `MYSQL_DATABASE` |\\n\\n### Backup & Restore\\n\\n**Identify the dump format:**\\n```bash\\nfile backup.dmp\\nhead -c 500 backup.dmp | cat -v\\n```\\n\\n- **SQL text** → restore with `psql` / `ksql` / `mysql`\\n- **Custom binary** → restore with platform tool (`pg_restore`, `sys_restore`, `mysqlrestore`)\\n\\n**Copy file into container:**\\n```bash\\ndocker cp backup.dmp db-container:/tmp/backup.dmp\\n```\\n\\n**Restore with --no-owner** (avoids \\\"role does not exist\\\" errors):\\n```bash\\n# PostgreSQL\\ndocker exec db-container pg_restore -U user -d dbname --no-owner /tmp/backup.dmp\\n\\n# KingbaseES\\ndocker exec db-container bash -c 'source /etc/profile && /path/to/sys_restore -U user -d dbname --no-owner /tmp/backup.dmp'\\n```\\n\\n**Handle \\\"already exists\\\" errors:**\\n```bash\\n# Terminate connections\\ndocker exec db-container psql -U user -d default_db -c \\\\\\n  \\\"SELECT pg_terminate_backend(pid) FROM pg_stat_activity WHERE datname = 'target_db' AND pid <> pg_backend_pid();\\\"\\n\\n# Drop and recreate\\ndocker exec db-container psql -U user -d default_db -c \\\"DROP DATABASE target_db;\\\"\\ndocker exec db-container psql -U user -d default_db -c \\\"CREATE DATABASE target_db;\\\"\\n\\n# Restore\\ndocker exec db-container pg_restore -U user -d target_db --no-owner /tmp/backup.dmp\\n```\\n\\n**Verify restore:**\\n```bash\\ndocker exec db-container psql -U user -d dbname -c \\\"\\\\dn\\\"\\ndocker exec db-container psql -U user -d dbname -c \\\\\\n  \\\"SELECT schemaname, count(*) FROM pg_tables WHERE schemaname NOT IN ('pg_catalog','information_schema') GROUP BY schemaname ORDER BY schemaname;\\\"\\n```\\n\\n### Connecting to the Database\\n\\nFrom host:\\n```bash\\npsql -h 127.0.0.1 -p 54321 -U username -d dbname\\n```\\n\\nFrom another container:\\n```bash\\ndocker run --rm -it --network host IMAGE psql -h 127.0.0.1 -p 54321 -U user -d db\\n```\\n\\n---\\n\\n## Part 4: MinIO Object Storage\\n\\n### Deploy MinIO\\n\\n```bash\\nsudo docker run -d \\\\\\n  --name minio \\\\\\n  -e MINIO_ROOT_USER=minioadmin \\\\\\n  -e MINIO_ROOT_PASSWORD=minioadmin \\\\\\n  -p 9000:9000 \\\\\\n  -p 9001:9001 \\\\\\n  -v minio_data:/data \\\\\\n  --restart unless-stopped \\\\\\n  minio/minio server /data --console-address \\\":9001\\\"\\n```\\n\\n- **Port 9000** — S3 API endpoint\\n- **Port 9001** — Web console (browser UI)\\n- **Volume** — `minio_data` for persistent storage\\n\\n### Create a bucket\\n\\nVia CLI (install `mc` first) or via the console at `http://localhost:9001`.\\n\\n### Connection from app\\n\\n```env\\nMINIO_ENDPOINT=\\\"http://localhost:9000\\\"\\nMINIO_ACCESS_KEY=\\\"minioadmin\\\"\\nMINIO_SECRET_KEY=\\\"minioadmin\\\"\\nMINIO_BUCKET=\\\"my-bucket\\\"\\nMINIO_USE_SSL=\\\"false\\\"\\n```\\n\\n---\\n\\n## Part 5: Full-Stack Deployment Pattern\\n\\nFor apps with database + object storage + app server (e.g., Next.js + Prisma + PostgreSQL + MinIO):\\n\\n1. **Docker services:** PostgreSQL, MinIO (and any other infra)\\n2. **Local app:** `npm install` → `npx prisma db push` → `npm run dev`\\n3. **`.env` file:** Point `DATABASE_URL` and `MINIO_ENDPOINT` to `localhost` (not Docker service names, since app runs on host)\\n\\nKey: Docker containers use `--name` for inter-container DNS, but host-to-container uses `localhost` + published ports.\\n\\n---\\n\\n## Part 6: GitHub Authentication for Cloning\\n\\nWhen `gh` CLI isn't available or token lacks `read:org` scope, use HTTPS + stored credentials:\\n\\n```bash\\ngit config --global credential.helper store\\n# Token stored in ~/.git-credentials after first use\\ngit clone https://USER:TOKEN@github.com/USER/REPO.git\\n```\\n\\n---\\n\\n## Pitfalls\\n\\n### P1: `sudo -S` + pipes break stdin\\n`echo 'pw' | sudo -S cmd1 | cmd2` — the second pipe consumes stdin, so sudo's password prompt may fail. **Fix:** chain commands with `&&` in a single `sudo -S sh -c '...'` or run each step as separate terminal calls.\\n\\n### P2: GPG dearmor fails with \\\"cannot open /dev/tty\\\"\\nWhen running `gpg --dearmor` non-interactively (via pipe/script), add `--batch` flag:\\n```bash\\nsudo gpg --batch --yes --dearmor -o /etc/apt/keyrings/docker.gpg /tmp/docker.gpg\\n```\\n\\n### P3: GPG \\\"dearmoring failed: File exists\\\"\\nIf `/etc/apt/keyrings/docker.gpg` already exists, gpg won't overwrite. Add `--yes` flag or `sudo rm` the file first.\\n\\n### P4: `tee` consumes stdin with sudo pipe\\n`echo 'pw' | sudo -S tee /path/to/file` works for the tee itself, but if you pipe content INTO it, the content pipe gets broken. **Fix:** write to `/tmp/file` first, then `sudo cp`.\\n\\n### P5: Docker repo not appearing in apt update\\nCheck `/etc/apt/sources.list.d/docker.list` exists and has correct content. The `tee` stdin issue (P4) often causes an empty file.\\n\\n### P6: \\\"permission denied\\\" on docker.sock despite user in docker group\\nAfter `usermod -aG docker $USER`, the group change doesn't take effect until a new login session.\\n\\n**Quick fix:**\\n```bash\\nsudo chmod 666 /var/run/docker.sock\\n```\\n\\n**Better fix** (persists across reboots):\\n```bash\\ngroups $USER   # verify 'docker' appears\\nsudo chown root:docker /var/run/docker.sock\\nsudo chmod 660 /var/run/docker.sock\\nnewgrp docker   # activate group without logout\\n```\\n\\n### P7: Docker Hub unreachable (IPv6 timeout or China network)\\nIf `docker pull` fails with timeout on IPv6 address, or China mirrors return 429/timeout:\\n- **Best option:** Configure proxy via systemd (see Part 2)\\n- Use mixed proxy port (e.g. 7890), not HTTP-only port\\n- Verify proxy works: `curl -x http://127.0.0.1:7890 -s https://registry-1.docker.io/v2/`\\n\\n### P8: \\\"repository does not exist or may require docker login\\\"\\nIf `docker pull` returns this error after proxy is confirmed working, the repo is private:\\n```bash\\ncurl -x http://127.0.0.1:7890 -s \\\"https://hub.docker.com/v2/repositories/OWNER/REPO/tags/?page_size=10\\\"\\n# {\\\"message\\\":\\\"object not found\\\"} → repo doesn't exist\\n# 401/403 → private repo, need: sudo docker login\\n```\\n\\n### P9: Kernel upgrade pending\\nDocker may warn about kernel version mismatch. Not blocking but recommend reboot when convenient.\\n\\n### P10: \\\"role does not exist\\\" during restore\\nThe dump references a role that doesn't exist in the target. Use `--no-owner` or create the role first.\\n\\n### P11: \\\"database is being accessed by other users\\\"\\nCan't drop a database with active connections. Terminate them first (see Backup & Restore section).\\n\\n### P12: Container starts but database not ready\\nSome databases take 10-30 seconds to initialize. Wait and check logs:\\n```bash\\nsleep 15 && docker logs db-container 2>&1 | tail -10\\n```\\n\\n### P13: Default password in image\\nMany database images have a hardcoded default password (often base64-encoded in the entrypoint). Always set your own via env vars.\\n\\n### P14: Encoding mismatch\\nIf restoring data with special characters (Chinese, etc.), ensure the target database encoding matches the source (e.g., `UTF-8`, `GBK`).\\n\\n### P15: NAT traversal for external access\\nWhen the app runs on an internal network and needs internet access:\\n- **Router port forwarding** is the most reliable\\n- **Cloudflare quick tunnels** (recommended, no registration): `cloudflared tunnel --url http://localhost:3000` → temporary `*.trycloudflare.com` HTTPS URL\\n- **ngrok** requires account + auth token; **localtunnel/serveo** often fail on restricted networks\\n- If the machine has a public IPv4 (check `curl -4 ifconfig.me`), port forwarding is simplest\\n- Always test external access: `curl -4 -s -o /dev/null -w \\\"%{http_code}\\\" http://PUBLIC_IP:PORT`\\n\\n---\\n\\n## Reference Files\\n\\n- `references/kingbasev8.md` — KingbaseES V8 specific notes (image, env vars, tools, restore workflow)\\n- `references/cloudflare-tunnels.md` — Expose local services to internet via Cloudflare tunnels\\n- `references/jellyfin-media-server.md` — Jellyfin media/TV streaming server deployment\\n\", \"path\": \"devops/docker/SKILL.md\", \"skill_dir\": \"/home/longshao/.hermes/skills/devops/docker\", \"linked_files\": {\"references\": [\"references/kingbasev8.md\", \"references/cloudflare-tunnels.md\", \"references/jellyfin-media-server.md\"]}, \"usage_hint\": \"To view linked files, call skill_view(name, file_path) where file_path is e.g. 'references/api.md' or 'assets/config.yaml'\", \"required_environment_variables\": [], \"required_commands\": [], \"missing_required_environment_variables\": [], \"missing_credential_files\": [], \"missing_required_commands\": [], \"setup_needed\": false, \"setup_skipped\": false, \"readiness_status\": \"available\"}",
      "tool_call_id": "call_df95c5969a39497aa6393820"
    },
    {
      "role": "tool",
      "content": "{\"success\": true, \"name\": \"hyperframes-deploy\", \"description\": \"Deploy and use HyperFrames — HTML-to-video rendering framework by HeyGen. Install, configure Chrome headless, preview, render MP4. Covers proxy pitfalls and onnxruntime issues.\", \"tags\": [\"video\", \"html\", \"rendering\", \"puppeteer\", \"ffmpeg\", \"heygen\", \"hyperframes\"], \"related_skills\": [], \"content\": \"---\\nname: hyperframes-deploy\\ndescription: \\\"Deploy and use HyperFrames — HTML-to-video rendering framework by HeyGen. Install, configure Chrome headless, preview, render MP4. Covers proxy pitfalls and onnxruntime issues.\\\"\\ntags: [video, html, rendering, puppeteer, ffmpeg, heygen, hyperframes]\\ntriggers:\\n  - hyperframes\\n  - html to video\\n  - html video rendering\\n  - heygen video\\n---\\n\\n# HyperFrames Deployment\\n\\nDeploy HeyGen's open-source HTML-to-video rendering framework. Write HTML compositions, preview in browser, render to MP4 via headless Chrome + FFmpeg.\\n\\n## Prerequisites\\n\\n- Node.js >= 22\\n- FFmpeg (system-installed)\\n- ~200MB disk for Chrome Headless Shell\\n\\n## Installation\\n\\n```bash\\nnpm install -g hyperframes --ignore-scripts\\n```\\n\\n**⚠️ Pitfall: `--ignore-scripts` is required.** The `onnxruntime-node` dependency's post-install script fails when a proxy is active (HTTP 302 redirect treated as error). Skipping scripts is safe — onnxruntime is only needed for AI features (background removal), not core preview/render.\\n\\n## Chrome Headless Shell Setup\\n\\n`hyperframes browser ensure` often fails behind proxies. Manual setup:\\n\\n```bash\\n# 1. Download\\nmkdir -p ~/.cache/hyperframes/chrome/chrome-headless-shell\\ncd ~/.cache/hyperframes/chrome/chrome-headless-shell\\ncurl -L -o chrome-headless-shell-linux64.zip \\\\\\n  \\\"https://storage.googleapis.com/chrome-for-testing-public/131.0.6778.85/linux64/chrome-headless-shell-linux64.zip\\\"\\n\\n# 2. Extract\\npython3 -c \\\"import zipfile; zipfile.ZipFile('chrome-headless-shell-linux64.zip').extractall('.')\\\"\\n\\n# 3. Place in expected path (critical!)\\nmkdir -p linux-131.0.6778.85/chrome-headless-shell-linux64\\ncp -r chrome-headless-shell-linux64/* linux-131.0.6778.85/chrome-headless-shell-linux64/\\nchmod +x linux-131.0.6778.85/chrome-headless-shell-linux64/chrome-headless-shell\\n```\\n\\n**⚠️ Pitfall: Path must be exact.** Puppeteer looks for `~/.cache/hyperframes/chrome/chrome-headless-shell/linux-<VERSION>/chrome-headless-shell-linux64/chrome-headless-shell`. If the path is wrong, render fails with \\\"Browser was not found at the configured executablePath\\\".\\n\\n## Project Init & Rendering\\n\\n```bash\\n# Create project\\nhyperframes init my-video\\ncd my-video\\n\\n# Preview (dev server, live reload)\\nhyperframes preview    # runs on http://localhost:3002\\n\\n# Render to MP4\\nhyperframes render     # output in renders/ directory\\n```\\n\\n## Key Concepts\\n\\n| Command   | What it does                            | Output       |\\n|-----------|----------------------------------------|--------------|\\n| `preview` | Dev server with live reload             | HTTP server  |\\n| `render`  | Headless Chrome frame capture + FFmpeg  | MP4 file     |\\n\\n- **preview and render can run simultaneously** — independent processes\\n- **render needs no visible browser** — uses headless Chrome in background\\n- **preview on headless server** — access via `http://server-ip:3002` from LAN or SSH tunnel\\n\\n## Composition Structure\\n\\n```html\\n<!doctype html>\\n<html>\\n<head>\\n  <script src=\\\"https://cdn.jsdelivr.net/npm/gsap@3.14.2/dist/gsap.min.js\\\"></script>\\n</head>\\n<body>\\n  <div id=\\\"root\\\" data-composition-id=\\\"main\\\" data-start=\\\"0\\\" data-duration=\\\"10\\\"\\n       data-width=\\\"1920\\\" data-height=\\\"1080\\\">\\n    <!-- Timed clips: class=\\\"clip\\\" + data-start + data-duration + data-track-index -->\\n    <div id=\\\"title\\\" class=\\\"clip\\\" data-start=\\\"0\\\" data-duration=\\\"5\\\" data-track-index=\\\"0\\\"\\n         style=\\\"font-size:128px; color:#fff; display:flex; align-items:center; justify-content:center; width:100%; height:100%; background:linear-gradient(135deg,#667eea,#764ba2);\\\">\\n      Hello World\\n    </div>\\n  </div>\\n  <script>\\n    window.__timelines = window.__timelines || {};\\n    const tl = gsap.timeline({ paused: true });\\n    tl.from(\\\"#title\\\", { opacity: 0, scale: 0.5, duration: 1 }, 0);\\n    window.__timelines[\\\"main\\\"] = tl;\\n  </script>\\n</body>\\n</html>\\n```\\n\\n**Rules:**\\n1. Every timed element needs `data-start`, `data-duration`, `data-track-index`\\n2. Visible timed elements **must** have `class=\\\"clip\\\"`\\n3. GSAP timelines must be `paused: true` and registered on `window.__timelines`\\n4. No `Date.now()`, `Math.random()`, or network fetches (deterministic rendering)\\n\\n## Proxy Pitfalls\\n\\nWhen behind a proxy (e.g., mihomo/Clash on port 7890):\\n\\n1. **npm proxy settings break puppeteer downloads** — unset before `hyperframes browser ensure`:\\n   ```bash\\n   npm config delete proxy\\n   npm config delete https-proxy\\n   unset HTTP_PROXY HTTPS_PROXY\\n   ```\\n2. **If npm proxy needed for other packages**, re-set after Chrome is installed\\n3. **onnxruntime-node always fails through proxy** — use `--ignore-scripts`\\n\\n## Quick Lint\\n\\n```bash\\ncd my-video\\nhyperframes lint        # check composition errors\\nhyperframes validate    # validate structure\\nhyperframes inspect     # detailed inspection\\n```\\n\", \"path\": \"creative/hyperframes-deploy/SKILL.md\", \"skill_dir\": \"/home/longshao/.hermes/skills/creative/hyperframes-deploy\", \"linked_files\": null, \"usage_hint\": null, \"required_environment_variables\": [], \"required_commands\": [], \"missing_required_environment_variables\": [], \"missing_credential_files\": [], \"missing_required_commands\": [], \"setup_needed\": false, \"setup_skipped\": false, \"readiness_status\": \"available\"}",
      "tool_call_id": "call_cae4c7e1f34a44f28dacada3"
    },
    {
      "role": "tool",
      "content": "{\"success\": true, \"name\": \"python-webapp-deploy\", \"description\": \"Deploy Python/Flask/Node.js web applications from source code. Covers venv setup, dependency installation, env config, missing asset handling, process management, npm CLI workarounds, SQLite config, and password reset.\", \"tags\": [\"devops\", \"python\", \"flask\", \"nodejs\", \"deployment\", \"webapp\", \"self-hosted\", \"sqlite\", \"npm\"], \"related_skills\": [\"docker\"], \"content\": \"---\\nname: python-webapp-deploy\\ndescription: \\\"Deploy Python/Flask/Node.js web applications from source code. Covers venv setup, dependency installation, env config, missing asset handling, process management, npm CLI workarounds, SQLite config, and password reset.\\\"\\nversion: 1.0.0\\nauthor: Hermes Agent\\nlicense: MIT\\nmetadata:\\n  hermes:\\n    tags: [devops, python, flask, nodejs, deployment, webapp, self-hosted, sqlite, npm]\\n    related_skills: [docker]\\ntriggers:\\n  - deploy python app\\n  - deploy flask app\\n  - deploy nodejs app\\n  - deploy web app from source\\n  - python webapp deployment\\n  - npm global install\\n  - sqlite config\\n  - self-hosted service deploy\\n  - service password reset\\n---\\n\\n# Web App Deployment (From Source)\\n\\nDeploy Python, Flask, FastAPI, and Node.js web applications from source code when Docker images aren't available. Covers the full workflow from clone to running service, including npm CLI workarounds and SQLite-based configuration.\\n\\n## When to Use\\n\\n- Deploying Flask/FastAPI/Django apps from source\\n- No Docker image available or Docker not preferred\\n- Need to modify source code during deployment\\n- Internal/intranet apps that aren't published to PyPI\\n\\n## Deployment Workflow\\n\\n### 1. Clone or Download Source\\n\\n```bash\\n# Standard git clone\\n# 1. Download via gh api (git clone/curl fail through proxy due to gnutls)\\nexport GH_TOKEN=ghp_xxxx\\ngh api repos/mifengac/multi-rider/tarball/feature/ai-analyst-rag > /tmp/multi-rider.tar.gz\\ncd /tmp && tar xzf multi-rider.tar.gz\\n# Dir name will be like: mifengac-multi-rider-<hash>/\\n```\\n\\n### 2. Create Virtual Environment\\n\\n```bash\\ncd <target-dir>\\npython3 -m venv .venv\\nsource .venv/bin/activate\\npython --version  # Verify\\n```\\n\\n**Pitfall:** Always use `python3 -m venv`, not `virtualenv`. System may have different Python versions.\\n\\n### 3. Install Dependencies\\n\\n```bash\\n# Standard\\npip install -r requirements.txt\\n\\n# With Chinese mirror (faster in China)\\npip install -i https://pypi.tuna.tsinghua.edu.cn/simple -r requirements.txt\\n\\n# If pip not in PATH, use uv\\nuv pip install -r requirements.txt\\n\\n# For projects with local wheel dependencies\\npip install --no-index --find-links ./wheels -r requirements.txt\\n\\n# For lockfile-based installs\\npip install -r requirements.lock\\n```\\n\\n**Pitfall:** Large deps (torch, torchvision) may timeout. Use mirrors and increase timeout. `pip install -q` reduces output noise.\\n\\n**Pitfall:** Some projects reference local `.whl` files in requirements.txt (e.g., `./wheels/clip-1.0-py3-none-any.whl`). Ensure the `wheels/` directory exists or remove those lines.\\n\\n### 4. Configure Environment Variables\\n\\nMost Python web apps use one of these patterns:\\n\\n```bash\\n# Pattern A: .env file in project root\\ncp .env.example .env\\n# Edit .env\\n\\n# Pattern B: app.env (loaded by app code)\\ncat > app.env << 'EOF'\\nFLASK_SECRET_KEY=change-me\\nAPP_HOST=0.0.0.0\\nAPP_PORT=5001\\nDATABASE_URL=postgresql://user:pass@host:5432/db\\nEOF\\n\\n# Pattern C: Export directly\\nexport FLASK_SECRET_KEY=change-me\\nexport APP_PORT=5001\\n```\\n\\n**Check the app's config loading code** to know which pattern it uses. Common locations:\\n- `shared/config/config.py`\\n- `config.py` or `settings.py`\\n- `.env.example` or `ops/app.env.example`\\n\\n### 5. Create Required Directories\\n\\nApps often expect directories to exist:\\n\\n```bash\\n# Check README or config for required dirs\\nmkdir -p output upload_tmp logs model datasets\\n```\\n\\n### 6. Initialize Database\\n\\n```bash\\n# Flask-Migrate / Alembic\\nflask db upgrade\\n\\n# Custom migration scripts\\npython scripts/run_migrations.py\\n\\n# Or just let the app init SQLite on first run\\n```\\n\\n### 7. Start the Application\\n\\n```bash\\n# Standard Flask\\npython app.py\\n\\n# With gunicorn (production)\\ngunicorn -w 4 -b 0.0.0.0:5001 app:app\\n\\n# Background with nohup\\nnohup python app.py > app.log 2>&1 &\\n\\n# As a background process (Hermes)\\n# Use terminal with background=true\\n```\\n\\n### 8. Verify\\n\\n```bash\\n# Health check\\ncurl -s http://localhost:<port>/healthz\\ncurl -s http://localhost:<port>/api/health\\n\\n# Main page\\ncurl -s http://localhost:<port>/ | head -20\\n```\\n\\n## Common Pitfalls\\n\\n### Missing Model/Asset Files\\n\\nMany ML-based apps don't ship model files in the repo (too large). Check:\\n- README for model download instructions\\n- `model/` directory structure requirements\\n- `.gitignore` for excluded paths\\n\\nThe app may start but return degraded health status. This is usually OK for initial testing.\\n\\n### Oracle/Database Driver Issues\\n\\nSome apps need Oracle Instant Client or other native drivers:\\n- Check for `instantclient_*` directories\\n- May need `ORACLE_IC_DIR` env var\\n- Thick mode vs thin mode for oracledb\\n\\n### Port Already in Use\\n\\n```bash\\n# Find what's using the port\\nlsof -i :5001\\n# or\\nss -tlnp | grep 5001\\n\\n# Kill or change port\\nexport APP_PORT=5002\\n```\\n\\n### Import Errors on Startup\\n\\nUsually means missing dependencies. Check:\\n- `requirements.txt` vs actual imports\\n- Local `wheels/` directory for offline deps\\n- System packages (e.g., `libgl1` for OpenCV)\\n\\n### Database Schema Mismatch After Migration\\n\\nWhen the app crashes with errors like `column \\\"X\\\" does not exist`, the migration scripts may not create all columns the code expects. This happens when:\\n- `scripts/sql/v*.sql` creates a partial schema\\n- `scripts/ddl_create_tables.sql` has the full schema but isn't run by the migration script\\n- Code references columns added in a later version\\n\\n**Diagnosis:**\\n```python\\nimport psycopg2\\nconn = psycopg2.connect(host='...', port=..., database='...', user='...', password='...')\\ncursor = conn.cursor()\\ncursor.execute(\\\"SELECT column_name FROM information_schema.columns WHERE table_schema='myschema' AND table_name='mytable' ORDER BY ordinal_position\\\")\\nprint([r[0] for r in cursor.fetchall()])\\n```\\n\\n**Fix — ALTER TABLE to add missing columns:**\\n```python\\ncursor.execute(\\\"ALTER TABLE myschema.mytable ADD COLUMN IF NOT EXISTS calc_time TIMESTAMP DEFAULT CURRENT_TIMESTAMP;\\\")\\nconn.commit()\\n```\\n\\n**Pitfall:** Always check both `scripts/sql/v*.sql` AND `scripts/ddl_create_tables.sql` — the DDL file often has the complete schema while migration files are incremental.\\n\\n### Tailwind CSS Not Built\\n\\nIf the app uses Tailwind and pages look broken:\\n```bash\\nnpm install\\nnpm run build:css\\n```\\n\\nThe compiled CSS is usually at `static/dist/tailwind.css`.\\n\\n## Node.js App Deployment\\n\\nWhen no Docker image is available and the app is Node.js-based:\\n\\n### npm Global Install + Direct Execution\\n\\nMany npm packages ship a CLI wrapper that may not work in all environments. The fallback is running the actual server directly:\\n\\n```bash\\n# 1. Install globally\\nnpm install -g <package-name>\\n\\n# 2. Try CLI first\\n<package-name> --help\\n\\n# 3. If CLI hangs/fails, find the actual entry point\\nls -la $(npm root -g)/<package-name>/\\n\\n# 4. Run directly with node\\ncd $(npm root -g)/<package-name>/app  # or appropriate dir\\nPORT=3000 HOSTNAME=0.0.0.0 node server.js\\n```\\n\\n**Why CLI wrappers fail:**\\n- Spawn child processes with detached TTY\\n- Auto-update checks that hang in non-interactive environments\\n- System tray mode detection fails on headless servers\\n- Missing SQLite native deps (need runtime bootstrap)\\n\\n## SQLite-Based App Configuration\\n\\nMany self-hosted apps store config in SQLite:\\n\\n```python\\nimport sqlite3, json\\n\\ndb_path = \\\"~/.<app>/db/data.sqlite\\\"\\nconn = sqlite3.connect(db_path)\\ncursor = conn.cursor()\\n\\n# List tables\\ncursor.execute(\\\"SELECT name FROM sqlite_master WHERE type='table';\\\")\\n\\n# Read settings\\ncursor.execute(\\\"SELECT * FROM settings;\\\")\\nrows = cursor.fetchall()\\nfor row in rows:\\n    data = json.loads(row[1])  # Usually JSON in second column\\n    print(json.dumps(data, indent=2))\\n```\\n\\n### Password Reset for bcrypt-Protected Apps\\n\\nWhen you don't know the password and can't access the UI:\\n\\n```python\\nimport sqlite3, bcrypt, json\\n\\ndb_path = \\\"~/.<app>/db/data.sqlite\\\"\\nconn = sqlite3.connect(db_path)\\ncursor = conn.cursor()\\n\\nnew_password = \\\"newpassword\\\"\\npassword_hash = bcrypt.hashpw(new_password.encode(), bcrypt.gensalt()).decode()\\n\\ncursor.execute(\\\"UPDATE settings SET data = ? WHERE id = 1;\\\",\\n               (json.dumps({\\\"password\\\": password_hash}),))\\nconn.commit()\\n```\\n\\n**Requires:** `pip install bcrypt` or `uv pip install bcrypt`\\n\\n### API-Based Proxy Configuration\\n\\nMany services expose proxy settings via REST API:\\n\\n```bash\\n# Login first\\ncurl -s -X POST http://localhost:<port>/api/auth/login \\\\\\n  -H \\\"Content-Type: application/json\\\" \\\\\\n  -d '{\\\"password\\\":\\\"<password>\\\"}' -c cookies.txt\\n\\n# Update proxy settings (PATCH often works better than PUT)\\ncurl -s -X PATCH -b cookies.txt http://localhost:<port>/api/settings \\\\\\n  -H \\\"Content-Type: application/json\\\" \\\\\\n  -d '{\\\"outboundProxyEnabled\\\":true,\\\"outboundProxyUrl\\\":\\\"http://127.0.0.1:7890\\\"}'\\n```\\n\\n**Pitfall:** Some APIs use PATCH for partial updates, not PUT. PUT may require all fields.\\n\\n### mihomo/Clash Proxy Node Management\\n\\n```bash\\n# List available proxies\\ncurl -s http://127.0.0.1:9090/proxies\\n\\n# Switch active node in a proxy group\\ncurl -s -X PUT http://127.0.0.1:9090/proxies/<group-name> \\\\\\n  -H \\\"Content-Type: application/json\\\" \\\\\\n  -d '{\\\"name\\\":\\\"<node-name>\\\"}'\\n\\n# Check node status (URL-encode the node name)\\ncurl -s \\\"http://127.0.0.1:9090/proxies/<url-encoded-name>\\\"\\n```\\n\\n**Pitfall:** Node names with emojis/special chars must be URL-encoded. Use Python's `urllib.parse.quote()`.\\n\\n## Updating an Existing Deployment\\n\\nWhen the app is already deployed and you need to pull latest code while preserving local config and data:\\n\\n```bash\\nPROJECT_DIR=/home/longshao/multi-rider-rag\\n\\n# 1. Backup local files that aren't in git\\ncp $PROJECT_DIR/.env /tmp/.env.bak 2>/dev/null\\ncp $PROJECT_DIR/app.env /tmp/app.env.bak 2>/dev/null\\ncp -r $PROJECT_DIR/.venv /tmp/.venv.bak 2>/dev/null\\ncp -r $PROJECT_DIR/model /tmp/model.bak 2>/dev/null\\n# Also backup: output/, face_data/, train_runs/, datasets/ if they have data\\n\\n# 2. Download new code (gh api works through gnutls-broken proxies)\\nexport GH_TOKEN=ghp_xxxx\\ngh api repos/owner/repo/tarball/branch > /tmp/repo.tar.gz\\ncd /tmp && tar xzf repo.tar.gz\\n\\n# 3. Replace code (keep .venv and data dirs)\\nrm -rf $PROJECT_DIR\\nmv /tmp/owner-repo-*/ $PROJECT_DIR\\n\\n# 4. Restore local files\\ncp /tmp/.env.bak $PROJECT_DIR/.env 2>/dev/null\\ncp /tmp/app.env.bak $PROJECT_DIR/app.env 2>/dev/null\\nmv /tmp/.venv.bak $PROJECT_DIR/.venv 2>/dev/null\\nrm -rf $PROJECT_DIR/model && mv /tmp/model.bak $PROJECT_DIR/model 2>/dev/null\\n\\n# 5. Install any new dependencies\\ncd $PROJECT_DIR && source .venv/bin/activate\\npip install -r requirements.txt\\n\\n# 6. Check for new migrations\\nls scripts/*.sql scripts/sql/*.sql 2>/dev/null\\n\\n# 7. Restart\\npkill -f \\\"python app.py\\\" 2>/dev/null\\n# Then start fresh (use terminal background=true in Hermes)\\n```\\n\\n**Pitfall:** When `pip install -r requirements.txt` fails because a local `.whl` file is missing (e.g., `./wheels/clip-1.0-py3-none-any.whl`), filter it out and install the rest:\\n```bash\\ngrep -v \\\"wheels/\\\" requirements.txt > /tmp/reqs-filtered.txt\\npip install -r /tmp/reqs-filtered.txt\\n```\\nThen source the missing wheel separately if the feature is actually needed.\\n\\n## Process Management\\n\\n### Check if Running\\n\\n```bash\\ncurl -s -o /dev/null -w \\\"%{http_code}\\\" http://localhost:<port>/healthz\\n```\\n\\n### Stop\\n\\n```bash\\n# Find PID\\npgrep -f \\\"python app.py\\\"\\n# Kill gracefully\\nkill <PID>\\n```\\n\\n### Restart\\n\\n```bash\\nkill $(pgrep -f \\\"python app.py\\\") 2>/dev/null\\ncd <project-dir> && source .venv/bin/activate && python app.py &\\n```\\n\\n### npm CLI tools hanging silently\\n- Symptom: `<tool> --help` produces no output, times out\\n- Cause: CLI wrapper spawns child process, checks for updates, or waits for TTY\\n- Fix: Find and run the actual server entry point directly (see Node.js section above)\\n\\n### SQLite database locked\\n- Symptom: `database is locked` error\\n- Cause: Another process has the DB open (the running service)\\n- Fix: Stop the service before modifying the DB\\n\\n### bcrypt not installed\\n- Symptom: `ModuleNotFoundError: No module named 'bcrypt'`\\n- Fix: `uv pip install bcrypt` (use `uv` if `pip` not in PATH)\\n\\n### API returns \\\"Unauthorized\\\"\\n- Most self-hosted services require authentication\\n- Check for default credentials (admin/admin, admin/password, admin/123456)\\n- If password unknown, reset via SQLite (see SQLite section above)\\n\\n### Proxy nodes showing \\\"alive\\\": false\\n- Node may be temporarily down or blocked\\n- Switch to another node in the same group\\n- Check mihomo logs: `~/.config/mihomo/mihomo.log`\\n\\n## Quick Checklist\\n\\n1. [ ] Clone/download source\\n2. [ ] Create venv: `python3 -m venv .venv && source .venv/bin/activate`\\n3. [ ] Install deps: `pip install -r requirements.txt`\\n4. [ ] Configure env vars (check config loading pattern)\\n5. [ ] Create required directories\\n6. [ ] Initialize database if needed\\n7. [ ] Start app: `python app.py`\\n8. [ ] Verify: `curl http://localhost:<port>/healthz`\\n9. [ ] Check for missing models/assets in health response\\n\\n## Reference Files\\n\\n- `references/multi-rider-example.md` — Flask + YOLO deployment with KingBase (real-world example)\\n- `references/schema-mismatch-and-models.md` — DDL vs migration schema mismatch fix pattern\\n- `references/nextjs-fullstack-deploy.md` — Next.js + Prisma + Docker infrastructure deployment\\n- `references/9router-deployment.md` — npm global CLI tool deployment (Node.js, SQLite config, proxy)\\n\", \"path\": \"devops/python-webapp-deploy/SKILL.md\", \"skill_dir\": \"/home/longshao/.hermes/skills/devops/python-webapp-deploy\", \"linked_files\": {\"references\": [\"references/multi-rider-example.md\", \"references/schema-mismatch-and-models.md\", \"references/nextjs-fullstack-deploy.md\", \"references/9router-deployment.md\"]}, \"usage_hint\": \"To view linked files, call skill_view(name, file_path) where file_path is e.g. 'references/api.md' or 'assets/config.yaml'\", \"required_environment_variables\": [], \"required_commands\": [], \"missing_required_environment_variables\": [], \"missing_credential_files\": [], \"missing_required_commands\": [], \"setup_needed\": false, \"setup_skipped\": false, \"readiness_status\": \"available\", \"metadata\": {\"hermes\": {\"tags\": [\"devops\", \"python\", \"flask\", \"nodejs\", \"deployment\", \"webapp\", \"self-hosted\", \"sqlite\", \"npm\"], \"related_skills\": [\"docker\"]}}}",
      "tool_call_id": "call_e62ab854d94b4509a31d23dd"
    },
    {
      "role": "assistant",
      "content": "## Assessment\n\nOnly **3 agent-created skills** exist. After reviewing each one's full content:\n\n| Skill | Category | Scope | Already umbrella? |\n|-------|----------|-------|--------------------|\n| `docker` | devops/infra | Install → configure → run databases/MinIO/full-stack | ✅ Yes — 6 parts, 15 pitfalls, 3 reference files |\n| `hyperframes-deploy` | creative/video | HeyGen HTML-to-video rendering | ✅ Yes — complete single-tool workflow |\n| `python-webapp-deploy` | devops/deploy | Python/Flask/Node.js source deployment | ✅ Yes — 9 steps, many pitfalls, 4 reference files |\n\n### Clustering analysis\n\n- **No prefix clusters exist.** The three skills share no first-word or domain-keyword prefix.\n- **`docker` ↔ `python-webapp-deploy`:** Both are devops/deployment, but they serve **different deployment paradigms** (containerized infra vs. source-code-to-running-service). `docker` already links to `python-webapp-deploy` via `related_skills`. The proxy overlap is contextual (Docker daemon proxy vs. npm/pip proxy) and belongs in each skill's own pitfalls section.\n- **`hyperframes-deploy`:** Completely distinct domain (video rendering framework). No overlap with the other two.\n- **All three are already class-level skills** with rich bodies, labeled subsections, pitfalls, and reference files. None has a narrow session-specific name (no PR numbers, error strings, or audit artifacts).\n\n### Decision\n\n**No consolidations or prunings.** All three skills are already well-structured class-level umbrellas with no meaningful overlap that merging would improve. With only 3 candidates and zero clusters, there's nothing to consolidate.\n\n## Structured summary\n```yaml\nconsolidations: []\nprunings: []\n```",
      "reasoning": "Let me analyze the three skills:\n\n1. **docker** - Docker on Linux: install Engine + Compose, configure proxy, run databases, MinIO, deploy full-stack apps. Category: devops.\n2. **hyperframes-deploy** - Deploy HyperFrames (HTML-to-video rendering). Category: creative/video.\n3. **python-webapp-deploy** - Deploy Python/Flask/Node.js web applications from source. Category: devops.\n\nNow let me think about clustering:\n\n- **docker** and **python-webapp-deploy** are both in the devops/deployment space. They share a deployment theme. docker covers containerized deployment, python-webapp-deploy covers source-based deployment. \n\nThey're related but serve different purposes — `docker` focuses on container setup, proxy configuration, and running databases and object storage, while `python-webapp-deploy` handles deploying apps from source code with venv and npm workflows. Even though they overlap on proxy configuration and deployment, they're distinct enough that I shouldn't try to merge them. The proxy knowledge in `python-webapp-deploy` is specific to npm and app contexts, not general Docker configuration.\n\nLooking at the actual candidates, `hyperframes-deploy` is too specialized (video rendering framework) to fit with the others, and all three skills are already well-structured and comprehensive. With only three candidates that don't share enough common ground, there's no meaningful clustering opportunity here. Looking at the actual content, `hyperframes-deploy` is too specialized to merge with the broader deployment skills, and `python-webapp-deploy` is already well-structured on its own. The real issue is just that `docker` has some generic proxy knowledge that overlaps with `python-webapp-deploy`'s proxy sections, but they're applied in different contexts. Given the instruction that \"keep\" is legitimate when a skill is already a class-level umbrella and merges wouldn't improve discoverability, I think all three should stay as-is. I've reviewed the three skills and determined they're each appropriately scoped for their domains—Docker for containerized infrastructure, HyperFrames for creative tooling, and Python webapp deployment for application deployment—so there's no meaningful consolidation to do here.",
      "finish_reason": "stop",
      "reasoning_content": "Let me analyze the three skills:\n\n1. **docker** - Docker on Linux: install Engine + Compose, configure proxy, run databases, MinIO, deploy full-stack apps. Category: devops.\n2. **hyperframes-deploy** - Deploy HyperFrames (HTML-to-video rendering). Category: creative/video.\n3. **python-webapp-deploy** - Deploy Python/Flask/Node.js web applications from source. Category: devops.\n\nNow let me think about clustering:\n\n- **docker** and **python-webapp-deploy** are both in the devops/deployment space. They share a deployment theme. docker covers containerized deployment, python-webapp-deploy covers source-based deployment. \n\nThey're related but serve different purposes — `docker` focuses on container setup, proxy configuration, and running databases and object storage, while `python-webapp-deploy` handles deploying apps from source code with venv and npm workflows. Even though they overlap on proxy configuration and deployment, they're distinct enough that I shouldn't try to merge them. The proxy knowledge in `python-webapp-deploy` is specific to npm and app contexts, not general Docker configuration.\n\nLooking at the actual candidates, `hyperframes-deploy` is too specialized (video rendering framework) to fit with the others, and all three skills are already well-structured and comprehensive. With only three candidates that don't share enough common ground, there's no meaningful clustering opportunity here. Looking at the actual content, `hyperframes-deploy` is too specialized to merge with the broader deployment skills, and `python-webapp-deploy` is already well-structured on its own. The real issue is just that `docker` has some generic proxy knowledge that overlaps with `python-webapp-deploy`'s proxy sections, but they're applied in different contexts. Given the instruction that \"keep\" is legitimate when a skill is already a class-level umbrella and merges wouldn't improve discoverability, I think all three should stay as-is. I've reviewed the three skills and determined they're each appropriately scoped for their domains—Docker for containerized infrastructure, HyperFrames for creative tooling, and Python webapp deployment for application deployment—so there's no meaningful consolidation to do here."
    }
  ]
}