About skill-compat-matrix
skill-compat-matrix is a repository skill that reports how skill metadata and bundled assets map to supported harnesses. It offers per-skill or repository-wide reports, including machine-readable JSON, without installing skills or changing harness configuration.
Key features
- 01Per-skill compatibility matrix
- 02Repository-wide compatibility summary
- 03Machine-readable JSON output
- 04Classification of portable and runtime-specific fields
- 05Compatibility reporting without configuration changes
Use cases
- 01Check harness compatibility before documenting a skill
- 02Review field portability across repository skills
- 03Generate a JSON compatibility report
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ wyattowalsh/ agents --skill skill-compat-matrixAfter installing, type/skill-compat-matrixin Claude Code to confirm it loaded.
Bundled files
References and scripts installed along with SKILL.md.
evals/evals.jsonscripts/check.pycompat_matrix.pyscripts/asset_toolkit/validate_hooks.pyDetails
From the project READMESkill Compat Matrix
Summarize how a skill's frontmatter and bundled assets map to supported harnesses.
Scope: Compatibility reporting only. Does not install skills or edit harness configs.
Dispatch
| $ARGUMENTS | Action |
|---|---|
<name> | Matrix for one skill |
repo / --all | Matrix summary for all repo skills |
--format json | Machine-readable report |
| Empty | Show harness list and field tiers |
Field Tiers
| Tier | Examples |
|---|---|
| portable-core | name, description, license, metadata |
| portable-but-variable | allowed-tools, scripts, references |
| runtime-specific | user-invocable, hooks, context, paths |
Load skills/skill-creator/references/runtime-compatibility.md when interpreting results.
Workflow
uv run python skills/skill-compat-matrix/scripts/compat_matrix.py <name>
uv run python skills/skill-compat-matrix/scripts/compat_matrix.py --all --format json
Validation Contract
uv run python skills/skill-compat-matrix/scripts/check.py
Critical Rules
- Mark unknown or best-effort harness behavior explicitly.
- Never treat runtime-specific fields as portable safety controls.
- Pair matrix output with skill-creator security audit for executable surfaces.
- Do not mutate harness projections from this skill.
- Include install-path notes only as documentation, not commands to run.
Comments
Markdown supported