CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

review-pr

Two-axis senior reviewer. Axis 1 (standards, "is it built right?") runs 6 specialized parallel sub-agents (Security, API/CRUD, Pattern enforcement, Simplicity, Reuse/DRY, Test hygiene) producing a severity-based report. Axis 2 (spec & behavior, "is it the right thing?") runs a 7th fresh-context sub-agent producing a behavior-change confirm-list for the human. Axes are never merged. Scoped to the current branch's diff against the default branch.

Quality

75%

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 ./.agents/skills/review-pr/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

77%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 exceptionally actionable, well-gated review protocol with exact commands, output contracts, and validation checkpoints — the sequencing and executability are near-ideal. Its weaknesses are redundancy (core rules restated multiple times) and a monolithic single-file structure where reference files would keep the body lean.

Suggestions

State the axes-never-merged rule and the mutation-adapter-skip rule once each (e.g., in the intro and in §5b) and reference them elsewhere instead of restating them in Agents 6, 7, §3, §5, and §6.

Move the AST01–10 agentic-surface audit checklist and the mutation-delta protocol into reference files (e.g., references/ast-audit.md, references/mutation-delta.md) with one-line pointers from Agent 1 and Agents 6/7, keeping SKILL.md as the overview.

DimensionReasoningScore

Conciseness

The bulk is dense, repo-specific operational policy (AST numbers, invariant citations, script paths) that Claude could not reconstruct, but there is real repetition: the axes-never-merged rule is restated roughly five times (intro, §3 header, §5, §5b, §6) and the mutation-adapter-skip rule three times (Agent 6, Agent 7, §5b), plus tone-of-voice coaching ('Professional, friendly, and collaborative') Claude does not need. Mostly efficient but could be tightened.

3 / 5

Actionability

Fully executable throughout: exact commands ('git merge-base origin/<default-branch> HEAD', 'git log --oneline <merge-base>..HEAD'), the precise posting endpoint ('gh api repos/{owner}/{repo}/pulls/{number}/reviews'), verbatim report templates in code blocks, the exact PR-description marker line, and named scripts to run. Copy-paste ready with concrete output contracts covering the common cases.

5 / 5

Workflow Clarity

A clearly sequenced protocol (§0–§7) with explicit checkpoints: a mandatory branch-scope lock before anything else, a fallback to the local default branch when no origin remote exists, exhaustive empty-state handling ('no findings is a valid verdict', '— none found.' sections), and a mandatory human-confirmation gate before the outward-facing batch PR post — so the batch-operation validation cap does not apply.

5 / 5

Progressive Disclosure

No bundle files exist (no references/, scripts/, or assets/), so everything lives in one ~300-line SKILL.md. Section headers and internal navigation are strong, but several blocks — the OWASP AST01–10 audit checklist, the mutation-delta protocol, and the §5/§5b report output contracts — are each substantial enough to be one-level-deep reference files rather than inline body content.

3 / 5

Total

16

/

20

Passed

Description

63%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 highly specific, well-differentiated description of what the skill does, undermined by the complete absence of when-to-use trigger guidance and of the natural vocabulary ('code review', 'pull request') users employ. The what is excellent; the when is missing, which caps its loading behavior.

Suggestions

Add an explicit trigger clause, e.g. 'Use when asked to review a pull request, do a code review of the current branch, or check a diff before merge.'

Include the natural trigger terms users actually say — 'code review', 'PR review', 'pull request', 'review my diff/branch' — so the skill surfaces on real requests.

Consider trimming internal jargon ('Two-axis', 'fresh-context sub-agent', 'confirm-list') from the first sentence in favor of user-facing language, keeping the structural detail as a follow-on.

DimensionReasoningScore

Specificity

The description enumerates multiple concrete actions: 'runs 6 specialized parallel sub-agents (Security, API/CRUD, Pattern enforcement, Simplicity, Reuse/DRY, Test hygiene) producing a severity-based report' plus a second axis producing 'a behavior-change confirm-list'. Coverage of what the skill does is comprehensive with no obvious gaps.

5 / 5

Completeness

The 'what' is clear and detailed, but there is no 'Use when...' clause or equivalent explicit trigger guidance; the 'when' is only weakly implied by 'Scoped to the current branch's diff against the default branch'. Per the rubric guideline, a missing Use-when clause caps completeness at 3.

3 / 5

Trigger Term Quality

Some relevant keywords are present ('reviewer', 'diff', 'branch', 'severity'), but the natural phrases a user would actually say — 'code review', 'PR review', 'pull request', 'review my changes' — are entirely absent, and the text leans on jargon like 'Two-axis', 'fresh-context sub-agent', and 'confirm-list'. More than 'a few natural terms missing', so it sits at the middle anchor rather than 4.

3 / 5

Distinctiveness Conflict Risk

The two-axis standards-vs-spec framing and the named sub-agent lenses carve a mostly distinct niche with only minor overlap risk against closely related code-review skills. Not 5 because distinct *triggers* are what is missing — without natural trigger phrases, a generic 'review this PR' request could as easily match another review skill.

4 / 5

Total

15

/

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

referenced_paths_exist

Referenced path issues: 2 possibly missing (scripts/behavior-delta.sh, scripts/guards.config.sh)

Warning

Total

15

/

16

Passed

Repository

agranado2k/tic-tac-toe-haskell

Path

.agents/skills/review-pr/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.