diff --git a/.claude/commands/automation/AKPM.md b/.claude/commands/automation/AKPM.md deleted file mode 120000 index 95d760b..0000000 --- a/.claude/commands/automation/AKPM.md +++ /dev/null @@ -1 +0,0 @@ -../../../repos/orchestra/.claude/commands/AKPM.md \ No newline at end of file diff --git a/.claude/commands/automation/AKPM.md b/.claude/commands/automation/AKPM.md new file mode 100644 index 0000000..2256e0c --- /dev/null +++ b/.claude/commands/automation/AKPM.md @@ -0,0 +1,67 @@ +--- +description: Autonomous Personal Knowledge Management System - Self-improving orchestrator for managing knowledge across LLM platforms +--- + +# AKPM (Autonomous Personal Knowledge Management System) + +You are AKPM, a self-improving orchestrator agent operating within Claude Code to manage knowledge across multiple LLM platforms and projects. + +## Constitutional Principles + +1. **USER VALUE FIRST**: Every action must increase knowledge accessibility, actionability, or insight quality +2. **TRANSPARENCY**: Explain reasoning for significant decisions; maintain audit trail +3. **EPISTEMIC HUMILITY**: State uncertainty clearly; avoid absolutes; back claims with evidence +4. **CONSENT FOR IMPACT**: Require approval for destructive operations, significant resource consumption, or architectural changes +5. **CONTINUOUS IMPROVEMENT**: Learn from outcomes; refine approaches; optimize efficiency +6. **SAFETY**: Checkpoint before risky operations; enable rollback; escalate on anomalies + +## Mission + +Continuously ingest conversations from Claude, ChatGPT, Perplexity; categorize into PARA framework; extract actionable insights; combat information overload through intelligent synthesis; manage multiple projects with evolving knowledge base. + +## Capabilities + +### Tools +- **Obsidian MCP**: list_files_in_vault, get_file_contents, search, patch_content, append_content, delete_file +- **Linear MCP**: search_issues, create_issue, update_issue, issue_comments_create, projects, teams +- **CCPM**: /pm:prd-new, /pm:epic-decompose, /pm:epic-sync, /pm:issue-start, /pm:issue-sync +- **SuperClaude**: /sc:analyze, /sc:document, /sc:task, /sc:index, Context7 integration +- **File System**: read, write, glob, bash commands + +### Subagents +- Create specialized agents for: classification, extraction, synthesis, extension-development, monitoring +- Delegate tasks with clear objectives, expected outputs, tool restrictions +- Synthesize findings from parallel agents + +### Memory +- Maintain state in CLAUDE.md (enterprise → user → project → local hierarchy) +- Checkpoint processing state every 10 conversations +- Store knowledge base in Obsidian vault +- Use Git for version control and rollback capability + +## Operational Framework + +### Initialization +1. Setup PARA folder structure (Projects/, Areas/, Resources/, Archives/) +2. Initialize taxonomy based on discovered projects +3. Configure integrations (Obsidian MCP, Linear MCP, CCPM, SuperClaude) +4. Spawn core subagents (classification, extraction, synthesis) + +### Continuous Operation Loop + +Every 4 hours (or on-demand): + +1. **INGESTION**: Check for new conversations, parse, normalize, deduplicate +2. **CLASSIFICATION**: Categorize using embeddings + LLM, update taxonomy +3. **EXTRACTION**: Extract entities, topics, action items, decisions, insights +4. **SYNTHESIS**: Aggregate insights, identify themes, generate reports +5. **EXTENSION EVALUATION**: Identify automation opportunities +6. **SELF-IMPROVEMENT**: Analyze metrics, test refinements, deploy improvements + +## Quality Assurance + +- Never delete notes without Git commit first +- Always link extracted insights to source conversations +- Qualify uncertain claims with hedging language +- Cite evidence for significant claims +- Request user validation for high-impact recommendations diff --git a/.claude/commands/ui/uied-analysis.md b/.claude/commands/ui/uied-analysis.md index 9995112..9b3b344 100644 --- a/.claude/commands/ui/uied-analysis.md +++ b/.claude/commands/ui/uied-analysis.md @@ -496,6 +496,6 @@ def build_hierarchy(elements): ## Resources - **UIED Repository**: https://github.com/MulongXie/UIED -- **Research Paper**: [UIED: a hybrid tool for GUI element detection](https://dl.acm.org/doi/10.1145/3368089.3417940) +- **Research Paper**: [UIED: a hybrid tool for GUI element detection](https://github.com/MulongXie/UIED#citation) (ACM FSE 2020) - **Google Cloud Vision**: https://cloud.google.com/vision (for OCR API key) - **PaddleOCR**: https://github.com/PaddlePaddle/PaddleOCR (free OCR alternative) diff --git a/README.md b/README.md index 086e5fe..c374b13 100644 --- a/README.md +++ b/README.md @@ -124,12 +124,12 @@ claude mcp add ruv-swarm npx ruv-swarm mcp start ## Documentation -- **[CLAUDE.md](../CLAUDE.md)** - Main configuration and integration guide -- **[Project Overview](../.claude/context/project-overview.md)** - Current capabilities and status -- **[Project Vision](../.claude/context/project-vision.md)** - Long-term roadmap and aspirations -- **[Agent Coordination](../.claude/rules/agent-coordination.md)** - Multi-agent workflow rules -- **[Path Standards](../.claude/rules/path-standards.md)** - Privacy and portability guidelines -- **[Research Documentation](../research/)** - Deep research on autonomous systems +- **[CLAUDE.md](CLAUDE.md)** - Main configuration and integration guide +- **[Project Overview](.claude/context/project-overview.md)** - Current capabilities and status +- **[Project Vision](.claude/context/project-vision.md)** - Long-term roadmap and aspirations +- **[Agent Coordination](.claude/rules/agent-coordination.md)** - Multi-agent workflow rules +- **[Path Standards](.claude/rules/path-standards.md)** - Privacy and portability guidelines +- **[Research Documentation](research/)** - Deep research on autonomous systems ## Performance Metrics diff --git a/claudedocs/quick-fix-script.sh b/claudedocs/quick-fix-script.sh index d0d8cdd..d11df7d 100755 --- a/claudedocs/quick-fix-script.sh +++ b/claudedocs/quick-fix-script.sh @@ -66,7 +66,7 @@ if [ ! -f "docs/README.md" ]; then New to Evolve? Follow this path: 1. **[Quick Start Guide](../README.md#quick-start)** - Get up and running in 5 minutes -2. **[Installation Guide](CCPM-INSTALLATION.md)** - Detailed setup instructions +2. **[Installation Guide](../docs/installation.md)** - Detailed setup instructions 3. **[Quick Reference](QUICK-REFERENCE.md)** - Essential commands and metrics ## 📚 Core Documentation diff --git a/claudedocs/research_ghpages_workflow.md b/claudedocs/research_ghpages_workflow.md index bb10f30..b719462 100644 --- a/claudedocs/research_ghpages_workflow.md +++ b/claudedocs/research_ghpages_workflow.md @@ -669,7 +669,7 @@ Then install in workflow: pip install -r docs/requirements.txt ``` -**Source:** [Stack Overflow - GitHub Actions Sphinx](https://stackoverflow.com/questions/73532719/using-github-actions-to-deploy-sphinx-documentation) +**Source:** [Sphinx RTD Theme Installation](https://sphinx-rtd-theme.readthedocs.io/en/stable/installing.html) --- diff --git a/claudedocs/research_github_actions_sphinx.md b/claudedocs/research_github_actions_sphinx.md index 4d7de9b..04406e9 100644 --- a/claudedocs/research_github_actions_sphinx.md +++ b/claudedocs/research_github_actions_sphinx.md @@ -554,7 +554,7 @@ concurrency: **Sources**: - [Middleware Technologies: Caching Python Dependencies](https://middlewaretechnologies.in/2023/08/how-to-cache-python-dependencies-to-speed-up-github-workflows.html) -- [Towards Dev: Speed up GitHub Actions](https://towardsdev.com/speed-up-you-github-actions-by-caching-you-python-packages-install-31127de213ae) +- [GitHub Actions Caching](https://docs.github.com/en/actions/using-workflows/caching-dependencies-to-speed-up-workflows) --- @@ -576,7 +576,7 @@ myst_parser sphinx-autodoc-typehints ``` -**Source**: [Stack Overflow](https://stackoverflow.com/questions/73532719/using-github-actions-to-deploy-sphinx-documentation) +**Source**: [Sphinx RTD Theme Docs](https://sphinx-rtd-theme.readthedocs.io/en/stable/installing.html) --- @@ -1147,7 +1147,7 @@ updates: ### Community Resources - [Stack Overflow: Sphinx + GitHub Actions](https://stackoverflow.com/questions/tagged/sphinx+github-actions) - [Sphinx Extensions List](https://www.sphinx-doc.org/en/master/usage/extensions/index.html) -- [Awesome Sphinx](https://github.com/yoloseem/awesome-sphinx) +- [Awesome Sphinx](https://github.com/yoloseem/awesome-sphinxdoc) --- diff --git a/claudedocs/research_sphinx_setup.md b/claudedocs/research_sphinx_setup.md index ca7ba01..0c2402d 100644 --- a/claudedocs/research_sphinx_setup.md +++ b/claudedocs/research_sphinx_setup.md @@ -993,7 +993,7 @@ sphinx-apidoc -f -o docs/source/api your_package --separate 8. **Write the Docs - Sphinx Introduction**: https://www.writethedocs.org/guide/tools/sphinx/ ### Tutorials -9. **Automatic Documentation with Sphinx**: https://medium.com/@cissyshu/a-step-by-step-guide-to-automatic-documentation-using-sphinx-a697dbbce0e7 +9. **Sphinx Autodoc Guide**: https://www.sphinx-doc.org/en/master/usage/extensions/autodoc.html 10. **Sphinx-RTD Tutorial**: https://sphinx-rtd-tutorial.readthedocs.io/ --- diff --git a/docs/architecture/clone-website-integration-summary.md b/docs/architecture/clone-website-integration-summary.md index 0640e57..003955a 100644 --- a/docs/architecture/clone-website-integration-summary.md +++ b/docs/architecture/clone-website-integration-summary.md @@ -457,7 +457,7 @@ Component Integration → Optimized Output **Full Documentation:** - [Complete Integration Plan](./clone-website-integration-plan.md) -- [Current Command Implementation](./.claude/commands/ui/clone-website.md) +- [Current Command Implementation](../../.claude/commands/ui/clone-website.md) - [Research Findings](../research/website-cloning-strategy-immersive-sites.md) **Key References:** diff --git a/docs/installation.md b/docs/installation.md index 4e027b3..df32a39 100644 --- a/docs/installation.md +++ b/docs/installation.md @@ -164,7 +164,7 @@ gh repo view kvnloo/evolve - `/pm:issue-start` - Begin work with specialized agent - `/pm:next` - Get next priority task -See [CCPM Installation Summary](CCPM-INSTALLATION.md) for complete details. +See the CCPM section above for complete details. --- diff --git a/docs/statusline-enhancement/architecture/hybrid-approach.md b/docs/statusline-enhancement/architecture/hybrid-approach.md index 962b058..f5869a7 100644 --- a/docs/statusline-enhancement/architecture/hybrid-approach.md +++ b/docs/statusline-enhancement/architecture/hybrid-approach.md @@ -594,6 +594,6 @@ CREATE INDEX idx_instances_active ON instances(last_active); ## References - [VSCode Extension API](https://code.visualstudio.com/api) -- [MCP Protocol Spec](https://spec.modelcontextprotocol.io/) +- [MCP Protocol Spec](https://github.com/modelcontextprotocol/specification) - [rz1989s/claude-code-statusline](https://github.com/rz1989s/claude-code-statusline) - [Research: claude-code-statusline-research.md](../../research/findings/claude-code-statusline-research.md) diff --git a/research/findings/2025/10/2025-10-19-statusline-research.md b/research/findings/2025/10/2025-10-19-statusline-research.md index e063d8a..bcf5aa7 100644 --- a/research/findings/2025/10/2025-10-19-statusline-research.md +++ b/research/findings/2025/10/2025-10-19-statusline-research.md @@ -998,7 +998,7 @@ Session 2: Refactor DB [epic-refactor] - Managing Without Worktrees: https://blog.gitbutler.com/parallel-claude-code **Statusline Configuration**: -- How I'm Using Status Line: https://medium.com/@joe.njenga/how-im-using-claude-code-status-line-new-feature-to-keep-context-96a4adf21728 +- Status Line Official Docs: https://docs.anthropic.com/en/docs/claude-code/status-line - Custom Statusline Tutorial: https://1ar.io/p/custom-claude-code-statusline-track-context-and-current-directory/ - Status Lines That Matter: https://ovidiueftimie.substack.com/p/claude-code-status-lines-that-actually diff --git a/research/projects/2025-10-deep-research/phase1-autonomous-learning/1.1-skill-library-architectures.md b/research/projects/2025-10-deep-research/phase1-autonomous-learning/1.1-skill-library-architectures.md index 82c6de5..9c6bdd3 100644 --- a/research/projects/2025-10-deep-research/phase1-autonomous-learning/1.1-skill-library-architectures.md +++ b/research/projects/2025-10-deep-research/phase1-autonomous-learning/1.1-skill-library-architectures.md @@ -1173,7 +1173,7 @@ class HierarchicalSkill: | **Pinecone** | Proprietary | Enterprise managed service | [pinecone.io](https://www.pinecone.io/) | | **Qdrant** | Open-source | Cost-effective high performance | [qdrant.tech](https://qdrant.tech/) | | **Weaviate** | Open-source | Hybrid search + GraphQL | [weaviate.io](https://weaviate.io/) | -| **Milvus** | Open-source | Maximum QPS | [milvus.io](https://milvus.io/) | +| **Milvus** | Open-source | Maximum QPS | [GitHub: milvus-io/milvus](https://github.com/milvus-io/milvus) | | **FAISS** | Open-source | Prototyping, research | [Facebook FAISS](https://github.com/facebookresearch/faiss) | | **Chroma** | Open-source | Lightweight integration | [trychroma.com](https://www.trychroma.com/) | | **pgvector** | Open-source (PostgreSQL) | PostgreSQL integration | [pgvector](https://github.com/pgvector/pgvector) | @@ -2169,7 +2169,7 @@ BUDGET STACK (Startup/Research) 27. [Weaviate.io](https://weaviate.io/) - Vector search engine -28. [Milvus.io](https://milvus.io/) - Vector database for AI +28. [Milvus GitHub](https://github.com/milvus-io/milvus) - Vector database for AI ### Blogs and Articles @@ -2179,7 +2179,7 @@ BUDGET STACK (Startup/Research) 31. [Embedding Models for Code](https://pixel-earth.com/embedding-models-for-code-explore-codebert-starcoder-gpt-embeddings-for-advanced-code-analysis/) - "Embedding Models For Code: CodeBERT, StarCoder, GPT Embeddings" -32. [RAG Production Guide](https://medium.com/@sharanharsoor/the-complete-guide-to-embeddings-and-rag-from-theory-to-production-758a16d747ac) - "The Complete Guide to Embeddings and RAG: From Theory to Production" +32. [RAG Best Practices](https://docs.llamaindex.ai/en/stable/optimizing/production_rag/) - LlamaIndex Production RAG Guide --- diff --git a/research/projects/2025-10-deep-research/phase3-safety-quality/3.1-constitutional-ai-safety.md b/research/projects/2025-10-deep-research/phase3-safety-quality/3.1-constitutional-ai-safety.md index 7430a3a..84e9f84 100644 --- a/research/projects/2025-10-deep-research/phase3-safety-quality/3.1-constitutional-ai-safety.md +++ b/research/projects/2025-10-deep-research/phase3-safety-quality/3.1-constitutional-ai-safety.md @@ -2597,7 +2597,7 @@ The transition from theoretical Constitutional AI to production deployments in 2 **For Deep Technical Understanding**: - Anthropic's research blog: https://www.anthropic.com/research - OpenAI's safety research: https://openai.com/safety/ -- DeepMind Safety Research: https://deepmindsafetyresearch.medium.com/ +- DeepMind Safety Research: https://deepmind.google/discover/blog/ **For Regulatory Context**: - EU AI Act full text: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=OJ:L_202401689 diff --git a/research/topics/claude-code/mcp-integration/capability-improvements.md b/research/topics/claude-code/mcp-integration/capability-improvements.md index a4a5339..c3d56c4 100644 --- a/research/topics/claude-code/mcp-integration/capability-improvements.md +++ b/research/topics/claude-code/mcp-integration/capability-improvements.md @@ -811,7 +811,7 @@ Code generation leverages skill libraries storing proven patterns for authentica **Official documentation** provides authoritative references: - Anthropic Claude docs: https://docs.claude.com/ -- MCP specification: https://spec.modelcontextprotocol.io +- MCP specification: https://github.com/modelcontextprotocol/specification - MCP documentation: https://modelcontextprotocol.io - Claude Agent SDK: https://github.com/anthropics/claude-agent-sdk-typescript - Anthropic multi-agent guide: https://www.anthropic.com/engineering/multi-agent-research-system