Code Search

Code Search Skill — Infrastructure Core

Best for low-token code understanding. Use before reading full files.

Role in OMP AEF:

• Enables semantic repository understanding for all framework operations
• Helps agents understand codebase structure and patterns
• Supports framework improvements and analysis
• Reduces token usage for large codebase navigation

Bash Commands

• generate_skeletons — Tree-sitter extraction of signatures/imports
• refresh_index — Rebuild vector embeddings (requires ollama)
• search_code(query, limit?) — Semantic search

Semantic search (via agent invocation)

task(role: "code-search", assignment: "Search for all milestone-related code")

Unified CLI usage (preferred)

bin/search_code "milestone" bin/search_code --index bin/search_code --skeletons

Capabilities

• Semantic repository search
• Tree-sitter skeleton generation
• Repository indexing

Preferred Usage

• Use before reading many source files.
• Prefer semantic search over exhaustive file reading.
• Prefer generated skeletons when architectural understanding is sufficient.
• Only inspect implementation files when additional detail is required.

Output

• Skeletons: docs/skeletons/{project}_skeleton.md
• Vector DB: code_index_{project}.db (per-project)

Requirements

• tree-sitter + tree-sitter-language-pack (installed)
• Optional: ollama + sqlite-vec for semantic search

For detailed usage, patterns, and integration with framework skills, see:

• code-search/README.md — Complete infrastructure documentation

Documentation

• skills.md — Comprehensive skill catalog
• INDEX.md — Complete skill catalog

References

• INDEX.md — Complete skill catalog
• AGENTS.md — Framework overview
• PLAYBOOK.md — Operational workflows
• FRAMEWORK.md — Architecture patterns