skill-graph
Skill1 starsChanged 7 months agoMITView on GitHub
Map how multiple SKILL.md files interact, conflict, or compose. Use when the user pastes a list of skills, a skills directory structure, or asks to audit a skill collection for redundancy, trigger-keyword overlap, or composition gaps. Returns a conflict matrix, redundancy flags, and recommended merges or splits.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- bryanwhl/skill-graphtheir own version, first seen May 6, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- bryanwhl/skill-graph★ 2same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository11 other files it gives its agents.
- assumption-surfaceSkill
- description-craftSkill
- goal-criteriaSkill
- karpathy-codingSkill
- lean-responseSkill
- simplicity-auditSkill
- skill-compressSkill
- surgical-diffSkill
- token-auditSkill
- token-budget-checkSkill
- vixero-councilSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 3 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 70
Do not rewrite skills. That is skill-compress's job.
- Distinctive: line 71
When trigger keywords conflict, recommend specificity in descriptions, not deletion of skills.
- Distinctive: line 72
If the collection has < 3 skills, return a lightweight report and suggest growing the collection before graphing.
- Distinctive: line 73
If two skills score 3 (redundant), pick the leaner one by token count to keep.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.
Files
skill-graph
Analyze a skills collection as a system. Find overlaps, gaps, and composition patterns.
Trigger on
- "audit my skills folder / collection"
- "do these skills conflict"
- "which skills overlap"
- User pastes two or more SKILL.md frontmatters together
- "how do my skills work together"
Input shapes accepted
- Multiple pasted SKILL.md files
- A list of
name: descriptionpairs - A directory listing (names only) — treat as partial analysis, warn about missing descriptions
Analysis passes
Pass 1 — trigger keyword extraction
For each skill, extract the trigger verbs and synonym nouns from its description.
Pass 2 — overlap matrix
Build a matrix. Two skills overlap when ≥ 2 trigger keywords match AND the stated output is similar. Score 0–3:
| score | meaning |
| 0 | disjoint |
| 1 | minor keyword overlap, different output |
| 2 | significant overlap — user intent ambiguous |
| 3 | redundant — merge or delete one |
Pass 3 — composition chains
Flag skills that naturally compose. If skill A's output is skill B's input, note it.
Example: token-audit → skill-compress → token-budget-check.
Pass 4 — coverage gaps
Identify intent categories the collection does not cover. Compare trigger verbs against common agentic task verbs (draft, review, audit, compose, generate, plan, debug, refactor, summarize, translate, explain, compare, rank).
Output
skills: {n}
total trigger keywords: {n}
unique trigger keywords: {n} ({ratio}% unique — higher is healthier)
overlap matrix (only scores ≥ 2 shown):
{skill-a} × {skill-b}: {score} — {reason}
...
composition chains:
{a} → {b} → {c}
{d} → {e}
redundancy flags:
MERGE: {a} + {b} (overlap 3)
SPLIT: {x} (does >1 job)
coverage gaps:
{missing-intent-1}
{missing-intent-2}Hard rules
- Do not rewrite skills. That is
skill-compress's job. - When trigger keywords conflict, recommend specificity in descriptions, not deletion of skills.
- If the collection has < 3 skills, return a lightweight report and suggest growing the collection before graphing.
- If two skills score 3 (redundant), pick the leaner one by token count to keep. Name both in the output.
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
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.

