CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

review-pr

Deep code review of a single open PR in nrwl/nx. Checks the PR out only inside an isolated sandbox, then runs four fixed reviewers: implementation (correctness, errors, types, performance), verification (tests, ticket grounding, comments, and docs), approach, and security. A reproduce-verifier executes a runnable repro only when verification identifies one. The skill saves a GitHub-flavored draft to ~/.nx-pr-reviews/<NUMBER>.md and never posts it. Claude reads/executes PR code only through the sandbox CLI; credentials never enter the sandbox.

Quality

74%

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

Quality

Content

70%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, rigorously validated orchestration workflow — every step is copy-paste executable with explicit checkpoints, retry loops, and failure recovery. Its weaknesses are the opposite trade-off: it is a ~1700-line monolith that is noticeably verbose (repeated trust-model statements, extensive war-story rationale) and keeps material — charter template, dispatch prompts, calibrations, grill protocol — inline that belongs in one-level-deep reference files.

Suggestions

Move the review-charter template, the agent dispatch prompts, and the Nx-specific calibration list into a references/ directory, leaving SKILL.md as a leaner overview with clearly signaled one-level-deep references.

Trim war-story rationales ('Observed: …' anecdotes, the six-bullet bug history of the verification block) to one-line comments, and state the sandbox trust model once with back-references instead of three times.

Cut the philosophical framing ('It is the exam-marking order', 'The independence that matters…') to their operative conclusions to reduce token load without losing the rules they justify.

DimensionReasoningScore

Conciseness

At ~1690 lines (~31k tokens) the body is noticeably verbose: the sandbox trust model is stated three times (the 'Trust model' section, Step 3's notes, and the 'Hard rule for every agent'), and long war-story rationales ('Observed: an image predating the pnpm-store warming went unnoticed for two weeks', the six-bullet history of past bugs in the EVIDENCE verification block, 'It is the exam-marking order') pad several sections beyond what the instruction needs. Not 1: none of it explains concepts Claude already knows — the verbosity is skill-specific operational rationale, not filler; not 3: the volume of narrative explanation and repetition is pervasive, not occasional.

2 / 5

Actionability

Fully executable throughout: literal bash blocks with fail-fast guards ('test -s /tmp/pr-<NUMBER>.diff.tmp || { echo "FATAL…"; exit 1; }'), complete copy-paste Agent() dispatch prompts, a single-verdict EVIDENCE verification script handling injection and newline edge cases, and a real bundled script (scripts/replay-prior-patch.sh) whose documented exit codes (10 on replay conflict, 20 on diff failure) match its actual code. Specific commands cover every common case from pre-flight to cleanup.

5 / 5

Workflow Clarity

Steps 1–10 are explicitly sequenced with validation checkpoints everywhere this batch/risky operation needs them: write-then-verify-then-move for host artifacts, changedFiles cross-check before dispatch, merge-base guards with FATAL aborts, a bounded one-retry feedback loop for unverified EVIDENCE lines, mandatory cleanup 'even on failure', and a dedicated 'On failure' recovery section. Branch conditions (force-push, replay fallback, noninteractive mode) are each handled explicitly.

5 / 5

Progressive Disclosure

Section structure is strong (numbered steps, clear headers, easy to navigate) and the one bundle reference — '${CLAUDE_SKILL_DIR}/scripts/replay-prior-patch.sh' — exists and is clearly signaled. But ~1700 lines are monolithic: large blocks that clearly belong in separate reference files (the ~230-line review-charter template, the dispatch prompts, the Nx-specific calibration list, the Step 8.5 grill protocol) are all inlined, and no references/ directory exists at all. Not 4: the bulk of detailed material is inline rather than split out; not 2: organization and navigation within the file are genuinely good, and the single reference that does exist is properly structured.

3 / 5

Total

15

/

20

Passed

Description

71%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, third-person, highly concrete description that clearly communicates what the skill does and its safety contract (sandbox isolation, drafts only, credentials walled off). Its main weakness is the complete absence of explicit trigger guidance — no 'Use when…' clause — leaving the 'when' implicit and capping completeness, and natural trigger-term coverage has minor gaps.

Suggestions

Add an explicit trigger clause, e.g. 'Use when asked to review, deep-review, or LGTM a pull request in the nrwl/nx repo', to raise completeness above 3.

Include the natural synonyms users say — 'pull request', 'PR review', 'review PR <number>' — so trigger matching covers common phrasings.

State the negative boundary explicitly ('not for PRs in other repos; not for posting reviews') to reduce overlap with generic code-review skills.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions with comprehensive coverage: 'Checks the PR out only inside an isolated sandbox, then runs four fixed reviewers: implementation (correctness, errors, types, performance), verification (tests, ticket grounding, comments, and docs), approach, and security', plus a conditionally-run 'reproduce-verifier', a concrete output path ('~/.nx-pr-reviews/<NUMBER>.md'), and an explicit non-behavior ('never posts it'). Not 4: coverage of what the skill does is complete, with no material gaps.

5 / 5

Completeness

The 'what' is clear and detailed, but there is no 'Use when…' clause or equivalent explicit trigger guidance anywhere in the description; the 'when' is only weakly implied by the subject matter. Per the judging guidelines, a missing 'Use when' clause caps completeness at 3. Not 4: the 'when' is entirely implicit, not merely imprecise.

3 / 5

Trigger Term Quality

Good natural keywords: 'Deep code review', 'open PR', 'code review' — phrases a user would plausibly say when invoking this. Not 5: common variations like 'pull request' (spelled out), 'review this PR', or 'LGTM' are missing, so a few natural trigger phrases go uncovered.

4 / 5

Distinctiveness Conflict Risk

A clear niche — 'a single open PR in nrwl/nx' with sandbox isolation and drafts-only output — distinguishes it from generic review skills, and phrases like 'sandbox CLI; credentials never enter the sandbox' are highly distinctive. Not 5: 'code review' and 'PR review' are trigger phrases that overlap any general code-review skill, so a user asking for a PR review in another repo could match this skill's terms. Not 3: the nrwl/nx scoping and draft-never-posts contract make real confusion unlikely.

4 / 5

Total

16

/

20

Passed

Validation

81%

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

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (1705 lines); consider splitting into references/ and linking

Warning

allowed_tools_field

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

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

13

/

16

Passed

Repository

nrwl/nx

Path

.claude/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.