{"skill":{"slug":"openclaw-knowledge-coach","displayName":"openclaw-knowledge-coach","summary":"Build and operate an OpenClaw-based local knowledge assistant that imports personal/local documents into a knowledge base and creates practice exercises duri...","description":"---\nname: openclaw-knowledge-coach\ndescription: Build and operate an OpenClaw-based local knowledge assistant that imports personal/local documents into a knowledge base and creates practice exercises during import. Use when users ask to set up OpenClaw knowledge workflows, ingest local notes/files, structure chunks and tags, or generate retrieval practice (quiz, flashcards, recall prompts) to master stored knowledge.\n---\n\n# OpenClaw Knowledge Coach\n\nCreate a local knowledge workflow in OpenClaw where importing knowledge also produces practice material for retention. OpenPraxis is on PyPI: use `pip install openpraxis` to get the `praxis` CLI.\n\n## CLI First\n\nUse OpenPraxis CLI as the default execution path.\n\n**Install from PyPI (recommended):**\n\n```bash\npip install openpraxis\npraxis --help\n```\n\nOr install from source for development:\n\n```bash\ngit clone https://github.com/Sibo-Zhao/OpenPraxis.git\ncd OpenPraxis\npip install -e \".[dev]\"\npraxis --help\n```\n\nConfigure provider/model/API key before ingestion/practice:\n\n```bash\npraxis llm setup\npraxis llm show\n```\n\nUse environment variables when needed (higher priority than config file):\n\n```bash\nexport OPENAI_API_KEY=\"your_key_here\"\n# or ARK_API_KEY / MOONSHOT_API_KEY / DEEPSEEK_API_KEY based on provider\n```\n\n## Core Workflow\n\n1. Confirm scope and source\n- Confirm knowledge domains, source folders, and accepted file types.\n- Confirm whether to preserve existing metadata (tags, dates, project names).\n\n2. Define import contract\n- Normalize each source into a record with `doc_id`, `title`, `source_path`, `tags`, `created_at`, and `content`.\n- Split long content into chunks with stable IDs such as `doc_id#chunk-001`.\n\n3. Import into OpenClaw\n- Ingest normalized records into the local OpenClaw knowledge base.\n- Keep a deterministic mapping between source file and imported IDs for later updates.\n\n4. Generate exercises at import time\n- For each chunk, create at least one retrieval exercise.\n- Prefer three exercise types:\n  - `free-recall`: ask the user to explain from memory.\n  - `qa`: ask direct question-answer pairs.\n  - `application`: ask scenario-based transfer questions.\n- Save answer keys and concise grading rubrics.\n\n5. Build review queue\n- Group exercises by topic and difficulty.\n- Schedule spaced review windows (for example: day 1, day 3, day 7, day 14).\n\n6. Validate quality\n- Reject exercises that can be answered without the imported knowledge.\n- Reject ambiguous or duplicate questions.\n- Ensure every exercise points back to `doc_id` and `chunk_id`.\n\n## CLI Command Playbook\n\nRun this sequence when the user asks to import local knowledge and create practice:\n\n1. Add a local file\n\n```bash\npraxis add \"/absolute/path/to/note.md\" --type report\n```\n\n2. List recent inputs and capture target `input_id`\n\n```bash\npraxis list --limit 20\n```\n\n3. Force-generate a new practice scene for an existing input\n\n```bash\npraxis practice <input_id>\n```\n\n4. Submit answer by file (preferred for deterministic runs)\n\n```bash\npraxis answer <scene_id> --file \"/absolute/path/to/answer.md\"\n```\n\n5. Inspect pipeline results and insight cards\n\n```bash\npraxis show <input_id>\npraxis insight <input_id>\n```\n\n6. Export insights to Markdown/JSON\n\n```bash\npraxis export --format md --output \"/absolute/path/to/insights.md\"\npraxis export --format json --output \"/absolute/path/to/insights.json\"\n```\n\n## Agent Execution Rules\n\n- Prefer `praxis add` for import and initial exercise generation.\n- Parse IDs from CLI output, then chain `praxis practice` and `praxis answer`.\n- Use `praxis answer --file` instead of interactive stdin in automation flows.\n- If duplicate content is skipped, rerun with `praxis add ... --force` when user wants reprocessing.\n- Use one-shot runtime model override only when requested:\n\n```bash\npraxis --provider openai --model gpt-4.1-mini add \"/absolute/path/to/note.md\"\n```\n\n- For image notes, pass image file path directly to `praxis add`; OCR extraction is built in.\n- Always finish with `praxis show` plus `praxis insight` or `praxis export` so user gets concrete output artifacts.\n\n## Output Contract\n\nWhen executing tasks with this skill, always provide these outputs:\n\n- Import summary: files processed, chunks created, failures.\n- Exercise summary: counts by type/topic/difficulty.\n- Review plan: next due batches and estimated workload.\n- Traceability map: `source -> doc_id -> chunk_id -> exercise_id`.\n\n## Exercise Format\n\nUse this compact JSON-like structure per exercise:\n\n```json\n{\n  \"exercise_id\": \"ex-...\",\n  \"doc_id\": \"...\",\n  \"chunk_id\": \"...\",\n  \"type\": \"free-recall | qa | application\",\n  \"question\": \"...\",\n  \"answer_key\": \"...\",\n  \"rubric\": [\"point 1\", \"point 2\"],\n  \"difficulty\": \"easy | medium | hard\",\n  \"next_review\": \"YYYY-MM-DD\"\n}\n```\n\nFor more generation patterns, read `references/exercise-patterns.md`.\n","tags":{"latest":"1.0.0"},"stats":{"comments":0,"downloads":654,"installsAllTime":0,"installsCurrent":0,"stars":2,"versions":1},"createdAt":1771806120426,"updatedAt":1778491612653},"latestVersion":{"version":"1.0.0","createdAt":1771806120426,"changelog":"OpenClaw Knowledge Coach 1.0.0 initial release\n\n- Launches a workflow for local document ingestion and automatic practice exercise generation in OpenClaw using OpenPraxis CLI.\n- Guides users through setup, import, normalization, and structured exercise creation tied to imported content.\n- Outlines a robust CLI command playbook for adding files, generating and answering practice, reviewing results, and exporting insights.\n- Introduces automated review planning with topic/difficulty grouping and spaced intervals.\n- Ensures output includes import summaries, exercise breakdowns, review schedules, and traceability from source file to exercise.\n- Standardizes exercise data format and enforces quality/relevance rules on generated practice material.","license":null},"metadata":null,"owner":{"handle":"sibo-zhao","userId":"s171zrjaph946s5q8rnwg6wrkn88484s","displayName":"Sibo-Zhao","image":"https://avatars.githubusercontent.com/u/128594101?v=4"},"moderation":null}