CtrlK
BlogDocsLog inGet started
Tessl Logo

local-review-codex

Run the CI Codex PR review locally against this branch's unpushed work (committed + uncommitted) before pushing. Same policy, model and reasoning effort as the codex-pr-review GitHub action.

70

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

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 exemplary instruction-style skill body: concrete commands, a clearly sequenced run workflow with error handling and fallbacks, and disciplined delegation of mechanics to run.sh. The only meaningful critique is slight redundancy in the cold-context rationale and time-sensitive version detail.

DimensionReasoningScore

Conciseness

The body is lean and operational — commands, a justified version pin, tight parity/difference lists — but the cold-context rationale is stated twice ('codex exec is a separate cold process...' and 'the value of a cold Codex pass is surfacing what the current session would rationalize away') and the version/upgrade detail is time-sensitive information outside any deprecated/old-patterns section, leaving minor trimmable excess short of anchor 5.

4 / 5

Actionability

Fully executable, copy-paste-ready commands cover the common cases: 'bash .agents/skills/local-review-codex/run.sh' with a custom-base variant, 'npm install --global @openai/codex@0.159.3', 'codex login', and 'git fetch', plus concrete fallback behavior (origin/main) and explicit relay instructions (print output verbatim).

5 / 5

Workflow Clarity

A clear Prerequisites → Run → Relay sequence with explicit checkpoints: the version check up front, the bash-vs-sh invocation caveat, the origin/main fallback, and the closing decision loop ('decide with the user whether to address findings before pushing'). The skill is read-only (no destructive/batch cap applies) and single-purpose with well-organized sections, satisfying the simple-skill exception for anchor 5.

5 / 5

Progressive Disclosure

No bundle files exist, and the body is under 50 lines with well-signaled sections (Correspondence with CI, Differences from CI, Prerequisites, Run, Relaying the result), delegating mechanics to run.sh instead of inlining them — matching the under-50-lines exception where well-organized sections alone warrant anchor 5.

5 / 5

Total

19

/

20

Passed

Description

78%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: third-person, concrete, and tightly scoped to its niche as the local pre-push twin of the codex-pr-review CI action. Its only gaps are a slightly integrated 'when' clause and a few missing natural trigger synonyms.

DimensionReasoningScore

Specificity

The description gives a concrete action with precise scope and timing — 'Run the CI Codex PR review locally against this branch's unpushed work (committed + uncommitted) before pushing' — but it is essentially one action with qualifiers rather than the multiple distinct actions of anchor 5.

4 / 5

Completeness

Both what ('Run the CI Codex PR review locally... (committed + uncommitted)') and when ('before pushing') are answered explicitly, but the when is embedded in the action sentence rather than a distinct 'Use when...' trigger clause, matching anchor 4's 'when could be more explicit' rather than anchor 5.

4 / 5

Trigger Term Quality

Natural keywords are present ('Codex', 'PR review', 'locally', 'unpushed', 'before pushing', 'GitHub action') and would be said by a user needing this skill, but a few natural variants like 'pre-push' or 'review my branch before push' are missing, so it falls short of anchor 5's synonym-level coverage.

4 / 5

Distinctiveness Conflict Risk

The description carves a clear niche — the local counterpart of one specific CI job ('Same policy, model and reasoning effort as the codex-pr-review GitHub action') — with distinct triggers ('CI Codex PR review', 'before pushing') and minimal conflict risk with generic review skills.

5 / 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
windmill-labs/windmill
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.