CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

git-repo-manager

Instructions to manage a local cache of GitHub repositories. This would typically done in cases where the user want to perform research/analysis on a repository. Invoke whenever you need to clone a repo that isn't present locally, bring an existing clone up to date, or remove a repo from the cache. This skill handles only the mechanical filesystem/git operations — not research, analysis, or anything about the repo's contents.

Quality

87%

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

87%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 exemplary lean, fully executable skill body whose commands are copy-paste ready and well organized. The one real weakness is the absence of validation around its destructive operation: 'rm -rf' is invoked with no verification step, capping workflow clarity.

Suggestions

Add a verification step before cleanup, e.g. confirm the target path matches 'temp/repo-cache/<name>' and exists before running 'rm -rf', since a mistyped name would delete the wrong directory.

Add a lightweight failure check after clone/pull (e.g. verify the directory now exists or the command exit status) so the workflow has an explicit checkpoint.

DimensionReasoningScore

Conciseness

Lean and efficient throughout: ~35 lines with no explanation of git or filesystem concepts Claude already knows, no padding, and every command earning its place. Matches the 5 anchor exactly.

5 / 5

Actionability

Fully executable, copy-paste-ready commands covering all three lifecycle cases: 'test -d temp/repo-cache/<name> && echo "exists" || echo "missing"', 'git clone --depth 1 <repo-url> temp/repo-cache/<name> --quiet', 'cd temp/repo-cache/<name> && git pull --quiet', and 'rm -rf temp/repo-cache/<name>'. Matches the 5 anchor with specific commands for the common cases.

5 / 5

Workflow Clarity

The sequence is clear (test existence, then clone if missing or pull if present; cleanup as a separate flow), but there are no validation checkpoints: no failure handling after clone/pull, and no verification before the destructive 'rm -rf temp/repo-cache/<name>'. Per the rubric guideline, missing validation for destructive operations caps this at 3, overriding the simple-skill exception. Not 4 because the destructive step has no guard whatsoever.

3 / 5

Progressive Disclosure

Under 50 lines, single-purpose, well-organized into clear sections (availability check, cleanup, rules) with no external references needed — no bundle files exist and none are referenced, so the structure is self-contained. Qualifies for 5 under the rubric's short-skill guideline.

5 / 5

Total

18

/

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 strong description that clearly states its niche (local repo cache lifecycle), enumerates concrete trigger conditions, and explicitly fences off research/analysis scope. Its main weaknesses are grammatical errors, second-person voice in the trigger clause, and a few missing natural synonyms.

Suggestions

Fix the grammar errors ('This would typically done in cases where the user want to...') and rephrase the trigger clause in third person to avoid the voice penalty, e.g. 'Invoke whenever a repo needs to be cloned locally, an existing clone updated, or a cached repo removed.'

Add common natural synonyms users would say — 'sync', 'update', 'mirror', 'check out' — to broaden trigger term coverage.

Tighten the redundant scope statements: the boundary is stated twice ('not research, analysis, or anything about the repo's contents' repeats the mechanical-operations framing); one clear statement would be leaner.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — 'clone a repo that isn't present locally, bring an existing clone up to date, or remove a repo from the cache' — which approaches the 5 anchor, but the second-person phrasing ('Invoke whenever you need to...') triggers the 1-point voice penalty, and grammar errors ('This would typically done', 'the user want') slightly blur the action statements.

4 / 5

Completeness

Explicitly answers both questions: what ('manage a local cache of GitHub repositories... handles only the mechanical filesystem/git operations') and when ('Invoke whenever you need to clone a repo that isn't present locally, bring an existing clone up to date, or remove a repo from the cache'), with an additional use context ('cases where the user want to perform research/analysis on a repository'). Matches the 5 anchor with concrete trigger phrases; the when-clause is fully explicit, not merely adequate.

5 / 5

Trigger Term Quality

Good natural keyword coverage: 'clone a repo', 'GitHub repositories', 'up to date', 'remove a repo from the cache'. Falls short of the 5 anchor because common synonyms users might say ('sync', 'fetch', 'mirror', 'update my clone') are absent.

4 / 5

Distinctiveness Conflict Risk

The explicit boundary clause ('not research, analysis, or anything about the repo's contents') carves a clear niche with minimal conflict against research/analysis skills. Not 5 because 'clone a repo' retains minor overlap risk with a general git or repo-setup skill.

4 / 5

Total

17

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

Total

15

/

16

Passed

Repository

coinbase/cds

Path

.claude/skills/git-repo-manager/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.