CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

git-master

MUST USE for ANY git operations: atomic commits, rebase/squash/fixup, and history search (blame, bisect, log -S, reflog). Load it before running git yourself, or pass load_skills=['git-master'] to the task that commits. Triggers: 'commit', 'rebase', 'squash', 'who wrote', 'when was X added', 'find the commit that'.

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

96%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.

Highly operational, lean content that assumes Claude's git competence while adding real procedural value: exact commands, verification checkpoints, recovery paths, and safety checklists for destructive operations. The only structural improvement is moving the peripheral PR-attachment upload protocol into a reference file.

DimensionReasoningScore

Conciseness

Every section is operational directives with concrete commands and zero explanation of concepts Claude already knows (it never explains what rebase or reflog are). Lines like "Be exact, conservative, and evidence-led" set tone in a single sentence — every token earns its place.

5 / 5

Actionability

Copy-paste-ready guidance throughout: the parallel Ground Truth command block, "GIT_SEQUENCE_EDITOR=: git rebase -i --autosquash <base>", "use --force-with-lease, never plain --force", "git blame -L start,end -- file", and exact HTTP endpoints with parameters for the attachment flow. Commands cover the common cases for each mode.

5 / 5

Workflow Clarity

The Mode Gate classifies requests first, Commit Mode is a numbered sequence with pre/post checkpoints ("Before each commit, verify git diff --staged --stat"), and Safety Checks is an explicit pre-write checklist with recovery paths ("use git rebase --abort first"). Validation for the destructive history-rewriting operations is explicit, so the destructive-operation cap at 3 does not apply.

5 / 5

Progressive Disclosure

Well-organized, self-contained sections with clear headers and no nested references (no bundle files exist), which is good structure. Not 5 because the skill exceeds 50 lines and the PR Body Evidence Attachments section — a GitHub upload protocol peripheral to core git operations — is a natural candidate for a one-level-deep reference file.

4 / 5

Total

19

/

20

Passed

Description

85%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 capability-dense description with concrete subcommand-level actions and an explicit trigger clause, hitting the top anchors for specificity and completeness. Its main weakness is the over-claiming "MUST USE for ANY git operations" scope, which raises conflict risk with other git-related skills.

Suggestions

Narrow "MUST USE for ANY git operations" to the actual domains covered (atomic commits, history rewriting, and history search) to reduce conflict with narrower git-related skills and avoid over-claiming.

Add natural trigger synonyms users actually say, such as "git blame", "git history", "amend", or "find when a line changed", to the trigger list to improve keyword coverage.

DimensionReasoningScore

Specificity

"atomic commits, rebase/squash/fixup, and history search (blame, bisect, log -S, reflog)" names multiple concrete operations by their actual git subcommands with comprehensive coverage across commit, rewrite, and search domains. Score 4 would apply only if there were minor gaps, but the capability list covers the skill's full scope.

5 / 5

Completeness

It explicitly answers what ("atomic commits, rebase/squash/fixup, and history search (blame, bisect, log -S, reflog)") and when (a literal "Triggers:" clause with concrete phrases). Both halves are explicit and concrete, matching the top anchor exactly.

5 / 5

Trigger Term Quality

Triggers "commit", "rebase", "squash", "who wrote", "when was X added", "find the commit that" are natural phrases users would say. Not 5 because common variations such as "blame", "git history", "amend", or "bisect" are missing from the trigger list even though they appear as capabilities.

4 / 5

Distinctiveness Conflict Risk

"MUST USE for ANY git operations" claims the entire git domain, creating real overlap risk with narrower git skills (commit-message generators, PR/GitHub helpers), though the distinctive subcommand niche (bisect, reflog, log -S) keeps it above a merely "very broad" skill. Not 4 because the "ANY" claim is more than minor overlap risk; not 2 because the concrete capability list anchors it to a specific niche.

3 / 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

code-yeongyu/oh-my-openagent

Path

packages/shared-skills/skills/git-master/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.