{ "schemaVersion": "1.0", "item": { "slug": "compound-mind", "name": "CompoundMind", "source": "tencent", "type": "skill", "category": "AI 智能", "sourceUrl": "https://clawhub.ai/cassh100k/compound-mind", "canonicalUrl": "https://clawhub.ai/cassh100k/compound-mind", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/compound-mind", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=compound-mind", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "brief.py", "clawpkg.yaml", "compound_mind.py", "data/distill_state.json", "data/experiences/039d83a0e521.json" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-30T16:55:25.780Z", "expiresAt": "2026-05-07T16:55:25.780Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network", "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/compound-mind" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/compound-mind", "agentPageUrl": "https://openagent3.xyz/skills/compound-mind/agent", "manifestUrl": "https://openagent3.xyz/skills/compound-mind/agent.json", "briefUrl": "https://openagent3.xyz/skills/compound-mind/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "CompoundMind v0.1", "body": "Makes agents permanently smarter. Each interaction compounds.\n\nThe problem: agents start from zero every session. Reading files helps, but raw logs are bulk. Real intelligence requires distillation - extracting what matters and making it instantly searchable." }, { "title": "What It Does", "body": "Distills memory logs into structured lessons, decisions, skill updates, relationships, and facts\nIndexes everything into a searchable SQLite database weighted by recency and importance\nBriefs you before any task with targeted lessons from past experience\nTracks growth over time - are you getting smarter or repeating mistakes?" }, { "title": "Quick Start", "body": "cd /root/.openclaw/workspace/compound-mind\n\n# Full pipeline (distill all memory files + build index)\npython compound_mind.py sync\n\n# Search your accumulated wisdom\npython compound_mind.py search \"Polymarket order types\"\npython compound_mind.py search \"git mistakes\" --category lesson\npython compound_mind.py search \"Chartist\" --category relationship\n\n# Pre-session briefing before a task\npython compound_mind.py brief \"trade on Polymarket BTC markets\"\npython compound_mind.py brief \"post content on X\"\npython compound_mind.py brief \"debug a Python cron job\"\n\n# Growth report\npython compound_mind.py report\n\n# Find repeated mistakes\npython compound_mind.py mistakes\n\n# Stats\npython compound_mind.py stats" }, { "title": "Commands", "body": "CommandWhat it doessyncDistill all new memory files + rebuild indexdistillExtract structured knowledge from memory filesrebuildRebuild the SQLite wisdom indexsearch Search accumulated wisdombrief Pre-session briefing for a specific taskreportGenerate growth report with LLM narrativemistakesShow repeated mistake patternsstatsIndex statistics" }, { "title": "File Structure", "body": "compound-mind/\n compound_mind.py - Main CLI\n distill.py - Experience distiller (uses Claude Haiku)\n index.py - SQLite FTS wisdom index\n brief.py - Pre-session briefing generator\n growth.py - Growth tracker and report generator\n data/\n experiences/ - Per-source distilled experience JSON files\n wisdom.db - SQLite FTS database\n growth.json - Growth tracking state\n briefs/ - Saved pre-session briefs\n distill_state.json - Tracks which files have been processed" }, { "title": "Distiller", "body": "Reads each memory file through Claude Haiku. Extracts:\n\nLessons - what worked, what failed, tagged by domain and outcome\nDecisions - context + action + outcome triplets\nSkill updates - evidence of capability improvement over time\nRelationships - how each person communicates, what they prefer\nFacts - specific knowledge worth keeping (wallet addresses, API patterns, config values)\n\nFiles are hash-tracked - re-runs only process changed files." }, { "title": "Wisdom Index", "body": "SQLite with FTS5 full-text search. Each entry scored by:\n\nFTS relevance (BM25)\nRecency (exponential decay, 30-day half-life)\nImportance (1-5 score assigned by distiller)" }, { "title": "Pre-Session Briefing", "body": "Given a task description, detects relevant domains, pulls top wisdom, synthesizes a sharp briefing via Claude Haiku. Covers:\n\nCritical lessons to remember\nPast failures to avoid\nKey facts and configs needed" }, { "title": "Growth Tracker", "body": "Analyzes all experience files to compute:\n\nLesson positive/negative ratios by domain\nDecision quality rates\nRepeated mistake patterns (same negative lesson appearing across multiple dates)\nComposite growth score (0-100)" }, { "title": "Integration with Agent Workflow", "body": "Ideal usage pattern:\n\nAfter each session - run sync (or schedule via cron)\nBefore each task - run brief \"task description\"\nWeekly - run report to see growth trajectory\nWhen stuck - run search \"relevant topic\" to surface past experience" }, { "title": "Cron Example (daily distillation)", "body": "0 3 * * * cd /root/.openclaw/workspace/compound-mind && python compound_mind.py sync --since $(date -d \"2 days ago\" +%Y-%m-%d) >> /tmp/compound-mind.log 2>&1" }, { "title": "Dependencies", "body": "Python 3.10+\nanthropic Python SDK (for distillation and briefing)\nSQLite3 (stdlib)\nMemory files at /root/.openclaw/workspace/memory/\n\nNo external databases. No vector embeddings. Runs entirely local with minimal API calls." }, { "title": "Design Principles", "body": "Incremental - only re-processes changed files\nCheap - uses Claude Haiku for extraction (low cost per memory file)\nFast - SQLite FTS5 for sub-second search\nHonest - growth tracking measures actual quality, not just quantity\nComposable - each module works standalone or as part of the pipeline" } ], "body": "CompoundMind v0.1\n\nMakes agents permanently smarter. Each interaction compounds.\n\nThe problem: agents start from zero every session. Reading files helps, but raw logs are bulk. Real intelligence requires distillation - extracting what matters and making it instantly searchable.\n\nWhat It Does\nDistills memory logs into structured lessons, decisions, skill updates, relationships, and facts\nIndexes everything into a searchable SQLite database weighted by recency and importance\nBriefs you before any task with targeted lessons from past experience\nTracks growth over time - are you getting smarter or repeating mistakes?\nQuick Start\ncd /root/.openclaw/workspace/compound-mind\n\n# Full pipeline (distill all memory files + build index)\npython compound_mind.py sync\n\n# Search your accumulated wisdom\npython compound_mind.py search \"Polymarket order types\"\npython compound_mind.py search \"git mistakes\" --category lesson\npython compound_mind.py search \"Chartist\" --category relationship\n\n# Pre-session briefing before a task\npython compound_mind.py brief \"trade on Polymarket BTC markets\"\npython compound_mind.py brief \"post content on X\"\npython compound_mind.py brief \"debug a Python cron job\"\n\n# Growth report\npython compound_mind.py report\n\n# Find repeated mistakes\npython compound_mind.py mistakes\n\n# Stats\npython compound_mind.py stats\n\nCommands\nCommand\tWhat it does\nsync\tDistill all new memory files + rebuild index\ndistill\tExtract structured knowledge from memory files\nrebuild\tRebuild the SQLite wisdom index\nsearch \tSearch accumulated wisdom\nbrief \tPre-session briefing for a specific task\nreport\tGenerate growth report with LLM narrative\nmistakes\tShow repeated mistake patterns\nstats\tIndex statistics\nFile Structure\ncompound-mind/\n compound_mind.py - Main CLI\n distill.py - Experience distiller (uses Claude Haiku)\n index.py - SQLite FTS wisdom index\n brief.py - Pre-session briefing generator\n growth.py - Growth tracker and report generator\n data/\n experiences/ - Per-source distilled experience JSON files\n wisdom.db - SQLite FTS database\n growth.json - Growth tracking state\n briefs/ - Saved pre-session briefs\n distill_state.json - Tracks which files have been processed\n\nHow It Works\nDistiller\n\nReads each memory file through Claude Haiku. Extracts:\n\nLessons - what worked, what failed, tagged by domain and outcome\nDecisions - context + action + outcome triplets\nSkill updates - evidence of capability improvement over time\nRelationships - how each person communicates, what they prefer\nFacts - specific knowledge worth keeping (wallet addresses, API patterns, config values)\n\nFiles are hash-tracked - re-runs only process changed files.\n\nWisdom Index\n\nSQLite with FTS5 full-text search. Each entry scored by:\n\nFTS relevance (BM25)\nRecency (exponential decay, 30-day half-life)\nImportance (1-5 score assigned by distiller)\nPre-Session Briefing\n\nGiven a task description, detects relevant domains, pulls top wisdom, synthesizes a sharp briefing via Claude Haiku. Covers:\n\nCritical lessons to remember\nPast failures to avoid\nKey facts and configs needed\nGrowth Tracker\n\nAnalyzes all experience files to compute:\n\nLesson positive/negative ratios by domain\nDecision quality rates\nRepeated mistake patterns (same negative lesson appearing across multiple dates)\nComposite growth score (0-100)\nIntegration with Agent Workflow\n\nIdeal usage pattern:\n\nAfter each session - run sync (or schedule via cron)\nBefore each task - run brief \"task description\"\nWeekly - run report to see growth trajectory\nWhen stuck - run search \"relevant topic\" to surface past experience\nCron Example (daily distillation)\n0 3 * * * cd /root/.openclaw/workspace/compound-mind && python compound_mind.py sync --since $(date -d \"2 days ago\" +%Y-%m-%d) >> /tmp/compound-mind.log 2>&1\n\nDependencies\nPython 3.10+\nanthropic Python SDK (for distillation and briefing)\nSQLite3 (stdlib)\nMemory files at /root/.openclaw/workspace/memory/\n\nNo external databases. No vector embeddings. Runs entirely local with minimal API calls.\n\nDesign Principles\nIncremental - only re-processes changed files\nCheap - uses Claude Haiku for extraction (low cost per memory file)\nFast - SQLite FTS5 for sub-second search\nHonest - growth tracking measures actual quality, not just quantity\nComposable - each module works standalone or as part of the pipeline" }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/cassh100k/compound-mind", "publisherUrl": "https://clawhub.ai/cassh100k/compound-mind", "owner": "cassh100k", "version": "0.1.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/compound-mind", "downloadUrl": "https://openagent3.xyz/downloads/compound-mind", "agentUrl": "https://openagent3.xyz/skills/compound-mind/agent", "manifestUrl": "https://openagent3.xyz/skills/compound-mind/agent.json", "briefUrl": "https://openagent3.xyz/skills/compound-mind/agent.md" } }