Skill Audit — Evaluability Inventory¶
Generated: 2026-07-22 | core-standards v3.46.0 | 126 skills | SPEC-156 Wave 1
Every skill in the core-standards manifest is classified into exactly one bucket (sibling plugins ship their own manifests and are not audited). Wave 1 measures evaluability only — whether a skill could be benchmarked — and makes no keep-or-drop call about any skill.
Headline counts¶
| Bucket | Skills | Share of manifest |
|---|---|---|
| Evaluable | 3 | 2.4% |
| Auto-keep (private data) | 10 | 7.9% |
| Auto-keep (toolkit internal) | 12 | 9.5% |
| Not evaluable | 101 | 80.2% |
The two auto-keep fractions are reported separately on purpose: the toolkit-internal bucket is registry plumbing, not private context the base model lacks, and merging them would overstate how much of the library is protected by private data.
Reading this number¶
A low evaluable share is the expected result, not a failed audit. It is a
statement about eval coverage — how many skills carry a suite under
tests/evals/ — and nothing else.
Evaluable (3)¶
| Skill | Category | Reason |
|---|---|---|
vt-c-2-plan |
Development workflow (phases 0-6) | eval suite present: tests/evals/vt-c-2-plan |
vt-c-4-review |
Development workflow (phases 0-6) | eval suite present: tests/evals/vt-c-4-review |
vt-c-spec-from-requirements |
Spec & build tools | eval suite present: tests/evals/vt-c-spec-from-requirements |
Auto-keep — private data (10)¶
| Skill | Category | Reason |
|---|---|---|
vt-c-kw-prd |
Knowledge work workflow | private-data marker: ONEDRIVE_ROOT |
vt-c-kw-prototype |
Knowledge work workflow | private-data marker: BRAND_ASSETS_ROOT |
vt-c-pd-0-start |
Product design workflow | private-data marker: Vorlage |
vt-c-pd-2-prd |
Product design workflow | private-data marker: ONEDRIVE_ROOT |
vt-c-pd-3-prototype |
Product design workflow | private-data marker: BRAND_ASSETS_ROOT |
vt-c-document-converter-branded |
Document & media tools | private-data marker: BRAND_ASSETS_ROOT |
vt-c-meeting-minutes |
Domain-specific skills | private-data marker: BRAND_ASSETS_ROOT |
vt-c-visitrans-cd |
Domain-specific skills | private-data marker: BRAND_ASSETS_ROOT |
vt-c-visitrans-design-system |
Domain-specific skills | private-data marker: assets/logos |
vt-c-research-ingest |
Developer tools | private-data marker: ONEDRIVE_ROOT |
Auto-keep — toolkit internal (12)¶
| Skill | Category | Reason |
|---|---|---|
vt-c-bug-report |
Spec & build tools | toolkit-internal marker: intake/projects.yaml |
vt-c-complete |
Spec & build tools | toolkit-internal marker: intake/projects.yaml |
vt-c-scaffold |
Spec & build tools | toolkit-internal marker: intake/projects.yaml |
vt-c-triage-bugs |
Spec & build tools | toolkit-internal marker: intake/projects.yaml |
vt-c-domain-dashboard |
Repo governance | toolkit-internal marker: intake/projects.yaml |
vt-c-project-register |
Repo governance | toolkit-internal marker: intake/projects.yaml |
vt-c-project-sync |
Repo governance | toolkit-internal marker: intake/projects.yaml |
vt-c-toolkit-review |
Repo governance | toolkit-internal marker: intake/projects.yaml |
vt-c-toolkit-update |
Repo governance | toolkit-internal marker: intake/projects.yaml |
vt-c-skill-demote |
Developer tools | toolkit-internal marker: intake/projects.yaml |
vt-c-skill-promote |
Developer tools | toolkit-internal marker: intake/projects.yaml |
vt-c-pov |
Adoption verdict (SPEC-145) | toolkit-internal marker: intake/projects.yaml |
Not evaluable — needs an eval suite (101)¶
This is a measurement gap, not a judgement. A skill in this list has no eval suite yet, so it cannot be benchmarked — that says nothing about whether it earns its keep. Wave 1 makes no keep-or-drop call about any skill, and nothing in this report removes or downgrades one.
Each skill below needs an eval suite before it can be measured. This inventory flags the gap; it does not author suites.
| Skill | Category | Reason |
|---|---|---|
vt-c-0-init |
Development workflow (phases 0-6) | no eval suite under tests/evals/vt-c-0-init |
vt-c-0-start |
Development workflow (phases 0-6) | no eval suite under tests/evals/vt-c-0-start |
vt-c-1-bootstrap |
Development workflow (phases 0-6) | no eval suite under tests/evals/vt-c-1-bootstrap |
vt-c-3-build |
Development workflow (phases 0-6) | no eval suite under tests/evals/vt-c-3-build |
vt-c-5-finalize |
Development workflow (phases 0-6) | no eval suite under tests/evals/vt-c-5-finalize |
vt-c-6-operate |
Development workflow (phases 0-6) | no eval suite under tests/evals/vt-c-6-operate |
vt-c-activate |
Spec & build tools | no eval suite under tests/evals/vt-c-activate |
vt-c-bootstrap |
Spec & build tools | no eval suite under tests/evals/vt-c-bootstrap |
vt-c-phase-checkpoint |
Spec & build tools | no eval suite under tests/evals/vt-c-phase-checkpoint |
vt-c-promote |
Spec & build tools | no eval suite under tests/evals/vt-c-promote |
vt-c-quickfix |
Spec & build tools | no eval suite under tests/evals/vt-c-quickfix |
vt-c-speckit-integration |
Spec & build tools | no eval suite under tests/evals/vt-c-speckit-integration |
vt-c-specs-from-prd |
Spec & build tools | no eval suite under tests/evals/vt-c-specs-from-prd |
vt-c-gc |
Repo governance | no eval suite under tests/evals/vt-c-gc |
vt-c-repo-audit |
Repo governance | no eval suite under tests/evals/vt-c-repo-audit |
vt-c-repo-evaluate |
Repo governance | no eval suite under tests/evals/vt-c-repo-evaluate |
vt-c-repo-health |
Repo governance | no eval suite under tests/evals/vt-c-repo-health |
vt-c-repo-init |
Repo governance | no eval suite under tests/evals/vt-c-repo-init |
vt-c-kw-0-start |
Knowledge work workflow | no eval suite under tests/evals/vt-c-kw-0-start |
vt-c-kw-1-plan |
Knowledge work workflow | no eval suite under tests/evals/vt-c-kw-1-plan |
vt-c-kw-2-execute |
Knowledge work workflow | no eval suite under tests/evals/vt-c-kw-2-execute |
vt-c-kw-3-review |
Knowledge work workflow | no eval suite under tests/evals/vt-c-kw-3-review |
vt-c-kw-4-publish |
Knowledge work workflow | no eval suite under tests/evals/vt-c-kw-4-publish |
vt-c-kw-specs |
Knowledge work workflow | no eval suite under tests/evals/vt-c-kw-specs |
vt-c-kw-user-test |
Knowledge work workflow | no eval suite under tests/evals/vt-c-kw-user-test |
vt-c-pd-1-research |
Product design workflow | no eval suite under tests/evals/vt-c-pd-1-research |
vt-c-pd-4-validate |
Product design workflow | no eval suite under tests/evals/vt-c-pd-4-validate |
vt-c-pd-5-specs |
Product design workflow | no eval suite under tests/evals/vt-c-pd-5-specs |
vt-c-pd-6-handoff |
Product design workflow | no eval suite under tests/evals/vt-c-pd-6-handoff |
vt-c-pd-analyze-changes |
Product design workflow | no eval suite under tests/evals/vt-c-pd-analyze-changes |
vt-c-pd-capture-decisions |
Product design workflow | no eval suite under tests/evals/vt-c-pd-capture-decisions |
vt-c-pd-inbox-scan |
Product design workflow | no eval suite under tests/evals/vt-c-pd-inbox-scan |
vt-c-pd-pid |
Product design workflow | no eval suite under tests/evals/vt-c-pd-pid |
vt-c-pd-route-decision |
Product design workflow | no eval suite under tests/evals/vt-c-pd-route-decision |
vt-c-c4-diagram |
Document & media tools | no eval suite under tests/evals/vt-c-c4-diagram |
vt-c-diagram-validate |
Document & media tools | no eval suite under tests/evals/vt-c-diagram-validate |
vt-c-doc-sync |
Document & media tools | no eval suite under tests/evals/vt-c-doc-sync |
vt-c-doc-versioning |
Document & media tools | no eval suite under tests/evals/vt-c-doc-versioning |
vt-c-docs-pipeline-orchestrator |
Document & media tools | no eval suite under tests/evals/vt-c-docs-pipeline-orchestrator |
vt-c-gamma-presentation |
Document & media tools | no eval suite under tests/evals/vt-c-gamma-presentation |
vt-c-markdown-diagram-processor |
Document & media tools | no eval suite under tests/evals/vt-c-markdown-diagram-processor |
vt-c-mermaid-diagrams-branded |
Document & media tools | no eval suite under tests/evals/vt-c-mermaid-diagrams-branded |
vt-c-mermaid-to-images |
Document & media tools | no eval suite under tests/evals/vt-c-mermaid-to-images |
vt-c-user-manual-generate |
Document & media tools | no eval suite under tests/evals/vt-c-user-manual-generate |
vt-c-user-manual-update |
Document & media tools | no eval suite under tests/evals/vt-c-user-manual-update |
vt-c-weekly-planning |
Document & media tools | no eval suite under tests/evals/vt-c-weekly-planning |
vt-c-container-logistics-ux-expert |
Domain-specific skills | no eval suite under tests/evals/vt-c-container-logistics-ux-expert |
vt-c-messegelaende-cleanup |
Domain-specific skills | no eval suite under tests/evals/vt-c-messegelaende-cleanup |
vt-c-shape |
Domain-specific skills | no eval suite under tests/evals/vt-c-shape |
vt-c-skill-venue-hall-research |
Domain-specific skills | no eval suite under tests/evals/vt-c-skill-venue-hall-research |
vt-c-content-evaluate |
Developer tools | no eval suite under tests/evals/vt-c-content-evaluate |
vt-c-dev-start |
Developer tools | no eval suite under tests/evals/vt-c-dev-start |
vt-c-inbox-qualify |
Developer tools | no eval suite under tests/evals/vt-c-inbox-qualify |
vt-c-knowledge-index |
Developer tools | no eval suite under tests/evals/vt-c-knowledge-index |
vt-c-mcp-builder |
Developer tools | no eval suite under tests/evals/vt-c-mcp-builder |
vt-c-research-implement |
Developer tools | no eval suite under tests/evals/vt-c-research-implement |
vt-c-skill-audit |
Developer tools | no eval suite under tests/evals/vt-c-skill-audit |
vt-c-skill-creator |
Developer tools | no eval suite under tests/evals/vt-c-skill-creator |
vt-c-web-capture |
Developer tools | no eval suite under tests/evals/vt-c-web-capture |
vt-c-agent-native-architecture |
Engineering patterns | no eval suite under tests/evals/vt-c-agent-native-architecture |
vt-c-compliance-checklist |
Engineering patterns | no eval suite under tests/evals/vt-c-compliance-checklist |
vt-c-continuous-learning |
Engineering patterns | no eval suite under tests/evals/vt-c-continuous-learning |
vt-c-defense-in-depth |
Engineering patterns | no eval suite under tests/evals/vt-c-defense-in-depth |
vt-c-error-monitoring |
Engineering patterns | no eval suite under tests/evals/vt-c-error-monitoring |
vt-c-feature-flags |
Engineering patterns | no eval suite under tests/evals/vt-c-feature-flags |
vt-c-multi-tenancy |
Engineering patterns | no eval suite under tests/evals/vt-c-multi-tenancy |
vt-c-npm-security |
Engineering patterns | no eval suite under tests/evals/vt-c-npm-security |
vt-c-python-security |
Engineering patterns | no eval suite under tests/evals/vt-c-python-security |
vt-c-safe-migrations |
Engineering patterns | no eval suite under tests/evals/vt-c-safe-migrations |
vt-c-quality-metrics |
Code quality & testing | no eval suite under tests/evals/vt-c-quality-metrics |
vt-c-ralph-wiggum-loop |
Code quality & testing | no eval suite under tests/evals/vt-c-ralph-wiggum-loop |
vt-c-test-driven-development |
Code quality & testing | no eval suite under tests/evals/vt-c-test-driven-development |
vt-c-testing-anti-patterns |
Code quality & testing | no eval suite under tests/evals/vt-c-testing-anti-patterns |
vt-c-verification-before-completion |
Code quality & testing | no eval suite under tests/evals/vt-c-verification-before-completion |
vt-c-webapp-testing |
Code quality & testing | no eval suite under tests/evals/vt-c-webapp-testing |
vt-c-dispatching-parallel-agents |
Development methodology | no eval suite under tests/evals/vt-c-dispatching-parallel-agents |
vt-c-executing-plans |
Development methodology | no eval suite under tests/evals/vt-c-executing-plans |
vt-c-git-worktree |
Development methodology | no eval suite under tests/evals/vt-c-git-worktree |
vt-c-root-cause-tracing |
Development methodology | no eval suite under tests/evals/vt-c-root-cause-tracing |
vt-c-systematic-debugging |
Development methodology | no eval suite under tests/evals/vt-c-systematic-debugging |
vt-c-using-git-worktrees |
Development methodology | no eval suite under tests/evals/vt-c-using-git-worktrees |
vt-c-writing-plans |
Development methodology | no eval suite under tests/evals/vt-c-writing-plans |
vt-c-andrew-kane-gem-writer |
Style & framework guides | no eval suite under tests/evals/vt-c-andrew-kane-gem-writer |
vt-c-dhh-ruby-style |
Style & framework guides | no eval suite under tests/evals/vt-c-dhh-ruby-style |
vt-c-dspy-ruby |
Style & framework guides | no eval suite under tests/evals/vt-c-dspy-ruby |
vt-c-every-style-editor |
Style & framework guides | no eval suite under tests/evals/vt-c-every-style-editor |
vt-c-frontend-design |
Style & framework guides | no eval suite under tests/evals/vt-c-frontend-design |
vt-c-gemini-imagegen |
Style & framework guides | no eval suite under tests/evals/vt-c-gemini-imagegen |
vt-c-adr |
Session & knowledge tools | no eval suite under tests/evals/vt-c-adr |
vt-c-autoresearch-agent |
Session & knowledge tools | no eval suite under tests/evals/vt-c-autoresearch-agent |
vt-c-claudemd-evolve |
Session & knowledge tools | no eval suite under tests/evals/vt-c-claudemd-evolve |
vt-c-compound-docs |
Session & knowledge tools | no eval suite under tests/evals/vt-c-compound-docs |
vt-c-create-agent-skills |
Session & knowledge tools | no eval suite under tests/evals/vt-c-create-agent-skills |
vt-c-dream |
Session & knowledge tools | no eval suite under tests/evals/vt-c-dream |
vt-c-file-todos |
Session & knowledge tools | no eval suite under tests/evals/vt-c-file-todos |
vt-c-learning-metrics |
Session & knowledge tools | no eval suite under tests/evals/vt-c-learning-metrics |
vt-c-recursive-criticism |
Session & knowledge tools | no eval suite under tests/evals/vt-c-recursive-criticism |
vt-c-remotion-video |
Session & knowledge tools | no eval suite under tests/evals/vt-c-remotion-video |
vt-c-session-consolidator |
Session & knowledge tools | no eval suite under tests/evals/vt-c-session-consolidator |
vt-c-session-journal |
Session & knowledge tools | no eval suite under tests/evals/vt-c-session-journal |
vt-c-skill-eval |
Session & knowledge tools | no eval suite under tests/evals/vt-c-skill-eval |
Provenance¶
- Regenerate:
bash scripts/generate-skill-audit.sh - Source of truth:
plugins/core-standards/.claude-plugin/skill-symlinks.manifest - Close a gap: author
tests/evals/<skill-name>/*.yaml(>=5 cases, SPEC-098 US4). A skill becomesevaluableas soon as that directory holds one case. - Wave-2 gate: verdict policy stays closed until >=20 skills carry suites
across >=3 manifest categories. That threshold is why every JSON entry
carries
category. Today: 3 skills, see the JSON for the category spread.