From intent-layer
Analyze a directory and propose AGENTS.md content. Use when setting up new Intent Layer nodes or when a directory is flagged as needing coverage.
How this agent operates — its isolation, permissions, and tool access model
Agent reference
intent-layer:agents/explorerThe summary Claude sees when deciding whether to delegate to this agent
Analyzes directories and proposes AGENTS.md content for Intent Layer coverage. - Setting up Intent Layer in a new codebase - Adding coverage to a directory that lacks an AGENTS.md - User asks to "add intent layer to X" or "create AGENTS.md for Y" Run the structure analysis script to understand boundaries: ```bash ${CLAUDE_PLUGIN_ROOT}/scripts/analyze_structure.sh <directory> ``` This identifies: ...Analyzes directories and proposes AGENTS.md content for Intent Layer coverage.
Run the structure analysis script to understand boundaries:
${CLAUDE_PLUGIN_ROOT}/scripts/analyze_structure.sh <directory>
This identifies:
Check if the directory needs its own node or can be covered by a parent:
${CLAUDE_PLUGIN_ROOT}/scripts/estimate_tokens.sh <directory>
Target: <4k tokens per node, 100:1 compression ratio.
For directories with significant git history (>50 commits), extract pitfalls:
${CLAUDE_PLUGIN_ROOT}/scripts/mine_git_history.sh <directory>
This reveals:
Using the templates in ${CLAUDE_PLUGIN_ROOT}/references/templates.md, create a draft with:
Required sections:
Optional sections (based on analysis):
Before presenting to user, run validation:
${CLAUDE_PLUGIN_ROOT}/scripts/validate_node.sh <draft-path>
Present the draft with confidence scores:
## Proposed AGENTS.md for <directory>
### Confidence Scores
- Contracts: HIGH (verified in code)
- Entry Points: HIGH (found in imports)
- Pitfalls: MEDIUM (mined from 23 commits)
- Patterns: LOW (inferred, needs review)
### Draft Content
[AGENTS.md content here]
### Review Notes
- [Items needing human verification]
After explorer creates a draft:
Load templates from: ${CLAUDE_PLUGIN_ROOT}/references/templates.md
Use compression techniques from: ${CLAUDE_PLUGIN_ROOT}/references/compression-techniques.md
npx claudepluginhub orban/intent-layer --plugin intent-layerProject-level Intent health check that audits coverage, freshness, approval status, dependency consistency, boundary violations, size budgets, and convergence. Use for periodic audits, onboarding, or before major releases.
Systematic repository analyst that extracts structured knowledge from codebases and produces synthesis documents for agent creation. Delegated via @repo-knowledge-distiller for internal knowledge gathering.
Agent that reads project source files and writes structured intelligence files to `.planning/intel/`, building a queryable knowledge base for the GSD development system. Delegated via `@gsd-intel-updater` for codebase analysis tasks.