CtrlK
BlogDocsLog inGet started
Tessl Logo

ce-commit

Create a git commit with a clear, value-communicating message. Use when the user asks to commit/save staged or unstaged changes with a repo-appropriate message.

72

Quality

88%

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

The canonical home for this skill is ce-commit in EveryInc/compound-engineering-plugin

SKILL.md
Quality
Evals
Security

Quality

Content

92%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.

An excellent, highly operational skill: exact commands, explicit stop and done conditions, validation checkpoints before and after the commit, and good/bad message examples. The only improvement space is mild tightening of two explanatory rationale paragraphs.

DimensionReasoningScore

Conciseness

The body is dense and operational throughout — it never explains git basics Claude already knows, and the Context table's "Non-zero / empty means" column is high-value compressed guidance. Two rationale paragraphs (why `git commit -F` avoids shell parsing, and why the trailing path list is required) each run longer than needed and could be trimmed to a sentence, matching anchor 4's "minor instances of over-explanation that could be trimmed" rather than anchor 5's every-token-earns-its-place.

4 / 5

Actionability

Every step is copy-paste ready: exact commands (`git checkout -b <name>`, `git commit -F <message-file> -- file1 file2 file3`, `git branch --show-current`), a fallback chain for the default branch, and good/bad message examples ("Fix double-submit on checkout" vs "Update checkout.rb"). Nothing is pseudocode or abstract; this matches the anchor-5 fully-executable pattern, and it is not level 4 because there are no gaps in the common-case path.

5 / 5

Workflow Clarity

A numbered 0–7 sequence with explicit validation checkpoints at every fragile point: step 1's nothing-to-commit stop (with the caveat that `git diff HEAD` misses untracked files), step 2's re-read of the branch after checkout, step 24's "Re-read branch and staged set immediately before committing", and step 7's post-commit `git status` confirmation. The "Done when / Stop when" line and the instruction to interpret non-zero exits rather than suppress them provide the checklist and error-recovery framing of anchor 5; the destructive/batch cap does not apply since verification is present.

5 / 5

Progressive Disclosure

This is a single-file skill of roughly 50 body lines with no references/, scripts/, or assets/ bundle, and none is needed — the skill never points to files that don't exist. Content is cleanly split into a Context section (command table) and a numbered Workflow, satisfying the simple-skill exception where well-organized sections alone earn anchor 5. It is not level 4 because there is no content that should have been split out into a separate file.

5 / 5

Total

19

/

20

Passed

Description

83%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.

A tight, well-formed description: concrete third-person action statement plus an explicit "Use when" clause with natural trigger synonyms. The only soft spots are single-action coverage in the "what" and the absence of the push/PR boundary in the description itself.

Suggestions

Consider enumerating one or two more concrete behaviors in the "what" (e.g., 'splits distinct concerns into logical commits') to move specificity from one action toward anchor 5's comprehensive coverage.

Add a clarifying boundary phrase like 'local commits only — not push or PR' to the description to reduce overlap risk with a full ship-flow skill at trigger time.

DimensionReasoningScore

Specificity

"Create a git commit with a clear, value-communicating message" names a concrete, precisely qualified action in correct third person. It stays at a single action rather than listing several specific behaviors (e.g., message authoring, branch handling, logical grouping), so it matches anchor 4 with minor coverage gaps rather than anchor 5's comprehensiveness.

4 / 5

Completeness

It explicitly answers both questions with concrete trigger phrases: the "what" is "Create a git commit with a clear, value-communicating message" and the "when" is an explicit "Use when the user asks to commit/save staged or unstaged changes with a repo-appropriate message" clause. This mirrors the anchor-5 example's structure; it cannot be the level below because anchor 4 requires a "when" that is weaker or less explicit, and there is nothing above.

5 / 5

Trigger Term Quality

"commit/save staged or unstaged changes" captures the natural synonyms users say ("commit", "save my changes", "staged") plus "repo-appropriate message" for the message angle. A few natural phrasings like "check in", "make a commit message", or "git commit" verbatim are absent, so it sits between anchor 4 and 5.

4 / 5

Distinctiveness Conflict Risk

The git-commit niche with commit/save triggers is well differentiated from generic skills, and "staged or unstaged changes" is a strong discriminator. There is minor overlap risk with a sibling commit-push-PR skill (whose boundary is only stated in the body as "No push, no PR — use ce-commit-push-pr"), keeping it just below anchor 5.

4 / 5

Total

17

/

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
crdant/compound-engineering-plugin
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.