CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

commit

Create a git commit for the Vision-Agents repo.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./.claude/skills/commit/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

This is a strong, disciplined skill body: it encodes only repo-specific constraints Claude could not infer, includes a real error-recovery loop for pre-commit hooks, and is organized into tight sections with no filler. The only meaningful gap is the absence of a concrete example commit message showing the expected conventional-commit shape, which would make the format rule fully copy-paste ready.

Suggestions

Add one example commit message (e.g. `feat(agents): add vision pipeline stage`) under the conventional-commits rule to make the expected format concrete.

DimensionReasoningScore

Conciseness

Every line is a repo-specific rule Claude would not otherwise know: "Keep the subject line ≤72 chars", "never `git add .` / `-A`", "Never pass `--no-verify`". There is zero padding, no explanation of what git or hooks are, and nothing that could be cut without losing information — a clean match for 'Lean and efficient; every token earns its place'.

5 / 5

Actionability

The guidance is concrete and directly executable — enforce conventional commits with a char limit, stage files by name, re-stage and retry when hooks modify files, fix root causes instead of bypassing hooks. It falls short of a 5 only because it states the format rule ("Use conventional commits") without a single example commit message to copy-paste, leaving a minor gap for an instruction-only skill.

4 / 5

Workflow Clarity

The flow is unambiguous for a simple single-purpose skill: check branch (warn on `main`), stage selectively, commit, then handle hook outcomes — and it includes an explicit feedback loop ("If a hook modifies files, re-stage and retry. If mypy or another check fails, fix the root cause before committing"). It is not a 5 because the sequence is never made explicit (the branch check, staging rules, and hook handling are scattered across sections rather than ordered), and not a 3 because validation/checkpoint behavior is clearly stated, not merely implicit.

4 / 5

Progressive Disclosure

The body is under 20 lines with no need for external references, and content is well-organized into two clearly signaled sections ("## Rules", "## Pre-commit hooks") that each hold exactly the right material. Per the rubric's simple-skill guidance, this is the case where well-organized sections alone merit a 5; nothing is inlined that belongs in a separate file.

5 / 5

Total

18

/

20

Passed

Description

53%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is concise, honest, and appropriately scoped to one repo and one task, with no fluff or over-claims. Its main weakness is the complete absence of trigger guidance — a user asking to "commit this" or "write a commit message" has no explicit 'Use when' signal — and it only gestures at one action rather than the skill's actual scope (branching, message format, hook handling). Adding a when-clause with natural trigger phrases would move it from adequate to strong.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user asks to commit, make a commit, or commit their changes in the Vision-Agents repo.'

Broaden natural trigger terms to include common variations like "commit my changes", "write a commit message", and "stage and commit".

Reflect the skill's real breadth in the description — mention conventional-commit message generation, selective staging, and pre-commit hook handling so the 'what' is comprehensive, not just one action.

DimensionReasoningScore

Specificity

"Create a git commit for the Vision-Agents repo" names a concrete domain and one concrete action (committing), but stops there — no mention of message generation, staging, branch handling, or hook interaction. This matches the anchor 'Names domain and 1-2 concrete actions, but not comprehensive'; a 4 would require several specific actions listed.

3 / 5

Completeness

The 'what' is clear (create a git commit for a specific repo), but there is no 'when' clause at all — no 'Use when...' or equivalent trigger guidance, which per the guidelines caps completeness at 3. It is not a 2 because the 'what' is fully clear, and not a 4 because 'when' is entirely absent rather than just imprecise.

3 / 5

Trigger Term Quality

"git commit" is the single natural keyword users would say, but common variations are absent — no "commit my changes", "make a commit", "commit this", or "write a commit message". This fits 'Some relevant keywords but missing common variations or synonyms'; a 4 would need broader natural-term coverage.

3 / 5

Distinctiveness Conflict Risk

Scoping to "the Vision-Agents repo" gives it a clear niche distinct from generic git/commit-message skills, leaving only minor overlap risk with closely related commit helpers. It is not a 5 because "create a git commit" alone could still compete with other commit-oriented skills outside that repo context.

4 / 5

Total

13

/

20

Passed

Validation

100%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

GetStream/Vision-Agents

Path

.claude/skills/commit/SKILL.md

Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.