CtrlK
BlogDocsLog inGet started
Tessl Logo

oma-scm

SCM (software configuration management) and Git — branching, merges, conflicts, worktrees, baselines, audit readiness, plus Conventional Commits and safe staging.

57

Quality

64%

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

Fix and improve this skill with Tessl

tessl review fix ./benchmarks/runs/oma/.agents/skills/oma-scm/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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.

A well-structured, safety-conscious skill with genuinely executable Git guidance and explicit approval guardrails for risky operations. Its main weaknesses are token inefficiency from re-teaching Conventional Commits basics Claude already knows, and a progressive-disclosure failure where detailed material is inlined while the referenced config/resource files are missing from the bundle.

Suggestions

Move the Conventional Commits tutorial (types table, format, Steps 1–5) into resources/conventional-commits.md and keep only the project-specific overrides (branch prefixes, 72-char limit, Co-Authored-By trailer) inline.

Ship or fix the referenced files: config/commit-config.yaml, config/cm-config.yaml, and the three resources/*.md files are cited but absent from the bundle, leaving guardrail step 2 and the References section unactionable.

Give the onboarding risk scan executable commands (e.g. git log --since / --numstat pipelines for churn and hotspot analysis) instead of bare metric names, and trim the boilerplate Scheduling subsections that restate the When to use / Guardrails content.

DimensionReasoningScore

Conciseness

The body re-teaches concepts Claude already knows — a full Conventional Commits table of feat/fix/refactor types, the "<type>(<scope>): <description>" format, and Steps 2–4 on choosing type/scope/description — plus a boilerplate "Scheduling" section (intent signature, expected inputs/outputs, control-flow features) that duplicates later content. It is not severely padded and does include genuinely non-obvious project rules, so it sits at 'mostly efficient but includes some unnecessary explanation' rather than a 2.

3 / 5

Actionability

Concrete, executable commands appear throughout: "git status -sb", "git diff --staged", the HEREDOC commit with a "-F /tmp/oma-commit-msg.txt" fallback, "git worktree add", "--force-with-lease", and "rerere", plus a crisp split rule ("one feature, few files (≤5)"). It is not a 5 because the onboarding risk scan ("High churn files in lookback window", "Bug hotspot files") gives no commands for computing these, and conflict handling ("resolve markers, test") stays high-level.

4 / 5

Workflow Clarity

Sequencing is explicit — Entry → Scenes (PREPARE…FINALIZE) → Transitions → Failure and recovery → Exit — with a VERIFY scene checking "status, staged diff, CI expectations, signatures, secrets", approval gates for history rewrites, and a CODEOWNERS checklist. It falls short of 5 because several checkpoints are labels rather than executable validation steps (e.g. "test" after conflict resolution names no command, and the onboarding scan has no verification procedure).

4 / 5

Progressive Disclosure

The body is well-sectioned but inlines substantial content that belongs in reference files (the full Conventional Commits tutorial, the onboarding risk scan, the CODEOWNERS playbook summary), and its references — "config/commit-config.yaml", "config/cm-config.yaml", "resources/conventional-commits.md", "resources/onboarding-risk-signals.md", "resources/codeowners-playbook.md", and "../../workflows/scm.md" — point to files that do not exist in this bundle (no references/, scripts/, or assets/ directories are present). That matches 'some structure but content that should be separate is inline' rather than the well-placed, clearly-navigable split of a 4.

3 / 5

Total

14

/

20

Passed

Description

66%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 domain-rich, specific description that clearly conveys what the skill covers, but it omits any explicit 'when to use' trigger guidance, which both caps completeness and weakens its retrieval function. Adding a 'Use when...' clause with natural user phrases would lift it substantially.

Suggestions

Append an explicit trigger clause, e.g. "Use when the user asks to commit, stage, branch, merge, rebase, resolve merge conflicts, manage worktrees, tag a release, or write Conventional Commit messages."

Include the natural verbs users actually say — "commit", "rebase", "tag/release", "pull request" — not just the noun forms ("merges", "staging").

Name the complementary skills it should not trigger for (e.g. debugging or code review) only if helpful, but keep the focus on concrete trigger phrases rather than domain labels.

DimensionReasoningScore

Specificity

The description lists several concrete capability areas — "branching, merges, conflicts, worktrees, baselines, audit readiness, plus Conventional Commits and safe staging" — going well beyond a generic domain label. It falls short of a 5 because these are topic nouns rather than concrete actions, and tagging/releases/committing are only implied rather than named.

4 / 5

Completeness

The 'what' is clearly stated (SCM/Git operations and governance), but there is no 'Use when...' clause or equivalent trigger guidance anywhere in the description — the 'when' is entirely absent, not just weakly implied. Per the judging guideline, a missing 'Use when' clause caps completeness at 3; a 4 would require at least an implicit or partial 'when'.

3 / 5

Trigger Term Quality

Natural keywords users would say are present: "Git", "branching", "merges", "conflicts", "worktrees", "staging", "Conventional Commits". A few common natural terms are missing ("commit", "rebase", "tag/release", "pull request"), matching the 'good keyword coverage; a few natural terms missing' anchor rather than the comprehensive synonym coverage of a 5.

4 / 5

Distinctiveness Conflict Risk

Distinctive terms like "audit readiness", "baselines", "Conventional Commits", and "safe staging" carve out a clear CM-governance niche. It stays at 4 rather than 5 because the broad "Git — branching, merges, conflicts" framing overlaps with generic Git or commit-message skills.

4 / 5

Total

15

/

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
first-fluke/oh-my-agent
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.