skill-graph-audit

Audit Skill() refs; detect hubs, isolates, and dangling targets

Night Market Skill — ported from claude-night-market/abstract. For the full experience with agents, hooks, and commands, install the Claude Code plugin.

Overview

Build a directed graph of Skill(plugin:name) invocations across the marketplace and surface composition patterns: which skills are heavily referenced (hubs), which orchestrate many others (orchestrators), which have no incoming or outgoing references (isolates), and which point at non-existent skills (dangling references).

The federation graph is now derivable from source rather than hand-curated.

When To Use

• Before a documentation pass on skill composition
• After a renaming or retirement to catch broken Skill() references
• During quarterly audits to spot orphaned skills
• When evaluating consolidation candidates (hubs are higher-risk to merge)
• When a new skill's outbound references should be sanity-checked

When NOT To Use

• For per-skill quality scoring -- use Skill(abstract:skills-eval) instead
• For frontmatter/structure validation -- use Skill(abstract:plugin-review)
• For hook-specific audits -- use Skill(abstract:hooks-eval)

Quick Start

bash

python3 plugins/abstract/scripts/skill_graph.py \
  --plugins-root plugins --top-n 10

For machine-readable output:

bash

python3 plugins/abstract/scripts/skill_graph.py \
  --plugins-root plugins --format json --output reports/skill-graph.json

See modules/usage.md for full CLI reference and example workflows.

Core Outputs

Output	Meaning	Action when high
Hubs	Most-referenced skills	Treat as core API; retire with extreme care
Orchestrators	Skills that call many others	Verify each ref still resolves
Isolates	Zero in / zero out	Check role: library? entrypoint? typo?
Dangling -- bugs	Missing internal target	Fix immediately (typo or retired skill)
Dangling -- external	Reference to external plugin	Document plugin dependency
Dangling -- placeholders	Template text like -NAME	Verify intentional

See modules/interpretation.md for false-positive guidance and isolation taxonomy.

Dogfood Evidence

This skill itself was scaffolded TDD-first; on first run against plugins/, it caught two genuine dangling refs that the manual audit (2026-04-25) had missed:

• attune:makefile-generation -> abstract:makefile-dogfooder (script name confused with skill name)
• imbue:karpathy-principles -> spec-kit:speckit-clarify (command referenced as skill)

Both were converted to correct command-style references in the same session.

Verification

Two ways to validate the audit output is trustworthy:

1. Test-suite correctness check: Run pytest -o addopts= plugins/abstract/tests/scripts/test_skill_graph.py to confirm extraction, graph construction, ranking, isolate detection, and dangling-ref classification all pass on the current code. The -o addopts= flag bypasses the package-wide coverage gate, which would otherwise fail on a single-file run.
2. Round-trip smoke check: Note the dangling-ref count from a baseline run, fix one or more flagged references, then rerun and verify the count drops by at least the number fixed. If the count does not move, the report is stale or the regex missed a syntax variant.

Related Skills

• Skill(abstract:skills-eval) -- per-skill quality scoring
• Skill(abstract:plugin-review) -- plugin manifest + structure
• Skill(abstract:hooks-eval) -- hook-specific validation
• Skill(abstract:rules-eval) -- rules directory validation

References

• Implementation: plugins/abstract/scripts/skill_graph.py
• Tests: plugins/abstract/tests/scripts/test_skill_graph.py
• Composition documentation: docs/quality-gates.md#skill-level-quality-gate-composition
• Skill role taxonomy: docs/skill-integration-guide.md#skill-role-taxonomy