agentleFS
clay-goodclay-good/spec-gen/skills/openlore-analyze-codebase/SKILL.md

openlore-analyze-codebase

Skill359 starsChanged 6 months agoMITView on GitHub

Run a full static OpenLore analysis and summarize architecture, call graph, refactoring issues, and duplicate code. Use when asked to analyze, map, or assess a codebase without LLM inference.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 5 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask the user which project to analyze, or confirm the current workspace root.

    line 23
  2. Very specific:

    Call the openlore MCP tool analyze_codebase with {"directory": "$DIRECTORY"}.

    line 34
  3. Very specific:

    If none of these files exist, skip this section and suggest running openlore analyze --force.

    line 52
  4. Distinctive:

    If no duplicates found, note this as a positive signal

    line 75
  5. Very specific:

    Call get_minimal_context on the highest-priority function — returns callers, callees, body, and test coverage in one call (~300 tokens).

    line 83
  6. Very specific:

    Call get_cluster on any function to see its full community (tightly coupled neighbors across directories).

    line 84
  7. Very specific:

    Call detect_changes to rank recently changed functions by blast radius — spot riskiest commits before reviewing.

    line 85
  8. Distinctive:

    If significant duplication was found, suggest consolidating clone groups before refactoring

    line 86
  9. Very specific:

    If the project has OpenSpec specs, call list_spec_domains then search_specs to enable spec-first reasoning (question → requirements → …

    line 88
  10. Distinctive:

    Never skip the duplication step — it determines the order of subsequent actions

    line 97
  11. Distinctive:

    Always present call graph and duplicate report results even if numbers are low

    line 98

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

openlore: Analyze Codebase

When to use this skill

Trigger this skill whenever the user asks to analyze a codebase with openlore, with phrasings like:

  • "analyze my project / my code"
  • "give me a code quality report"
  • "what are the structural issues in my codebase"
  • "find duplicates in my code"
  • explicit command /openlore-analyze-codebase

This skill is read-only — it modifies no files. It produces a report and suggests next steps.


Step 1 — Confirm the project directory

Ask the user which project to analyze, or confirm the current workspace root.

Which project directory should I analyze?
Options: current workspace root | enter a different path

Step 2 — Run static analysis

Call the openlore MCP tool analyze_codebase with {"directory": "$DIRECTORY"}.


Step 3 — Summarize the results

Present a concise summary:

  • Project type and detected frameworks
  • File count, function count, internal call count
  • Top 5 refactoring issues (function name, file, issue type, priority score)
  • Detected domains

Also report stack inventory (read directly from .openlore/analysis/ — no extra MCP call needed):

  • HTTP routes: N routes across M files — if route-inventory.json exists
  • ORM tables: N tables — if schema-inventory.json exists
  • Env vars: N total, X required without default — if env-inventory.json exists
  • UI components: N components — if ui-inventory.json exists

If none of these files exist, skip this section and suggest running openlore analyze --force.


Step 4 — Show the call graph

Call the openlore MCP tool get_call_graph with {"directory": "$DIRECTORY"}.

Highlight:

  • Hub functions (fanIn ≥ 8) — over-solicited functions, high coupling risk
  • Layer violations detected (e.g. a UI layer calling the database directly)

Step 5 — Show duplicate code report

Call the openlore MCP tool get_duplicate_report with {"directory": "$DIRECTORY"}.

Present a concise summary:

  • Overall duplication ratio (e.g. "12% of functions are duplicated")
  • Top 3 clone groups sorted by impact (instances × line count):
    • Clone type (exact / structural / near) and similarity score
    • List of instances (file + function name + line range)
  • If no duplicates found, note this as a positive signal

Step 6 — Suggest next steps

Based on the analysis, guide the user through the natural next actions in order:

  1. Call get_minimal_context on the highest-priority function — returns callers, callees, body, and test coverage in one call (~300 tokens). Use instead of get_subgraph + get_signatures separately.
  2. Call get_cluster on any function to see its full community (tightly coupled neighbors across directories).
  3. Call detect_changes to rank recently changed functions by blast radius — spot riskiest commits before reviewing.
  4. If significant duplication was found, suggest consolidating clone groups before refactoring
  5. Suggest running /openlore-plan-refactor once the user has enough context to act, then /openlore-execute-refactor to apply the plan
  6. If the project has OpenSpec specs, call list_spec_domains then search_specs to enable spec-first reasoning (question → requirements → linked source files). To activate search_specs, run openlore analyze --embed or openlore analyze --reindex-specs.

Absolute constraints

  • No code modifications in this workflow
  • Never skip the duplication step — it determines the order of subsequent actions
  • Always present call graph and duplicate report results even if numbers are low
  • Next steps (Step 6) are suggestions, not automatic actions

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.