CtrlK
BlogDocsLog inGet started
Tessl Logo

review-pr

Perform a review on a GitHub PR, leaving comments on the PR

56

Quality

71%

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

78%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.

The content is a well-structured, executable runbook: prerequisites with validation, exact gh commands for every step, and a clearly split posting format (inline comments plus a summary). Its main weaknesses are minor redundancy, a mismatch between the instructed `gh api` call and the frontmatter allowed-tools, and the absence of any post-comment verification step.

Suggestions

Add `Bash(gh api:*)` (or the specific POST form) to allowed-tools so the instructed inline-comment step in Part A is actually permitted.

Add a light verification step after posting, e.g. re-run `gh pr view <number> --json comments` or check each command's exit status before moving on, to close the workflow validation gap.

Trim the redundant lines ('Post inline comments using the gh api call', 'Keep the summary brief; put details in inline comments') since Part A/Part B already specify these.

DimensionReasoningScore

Conciseness

The body is lean and imperative with no explanations of concepts Claude already knows (no preamble about what code review is, no library tutorials). It is not a 5 because of small redundancies: 'Post inline comments using the gh api call' and 'Keep the summary brief; put details in inline comments' restate what Part A/Part B already specify. Not a 3 — padding is minimal and nearly every line carries instructions.

4 / 5

Actionability

Guidance is highly concrete: exact commands are given for every step (`gh pr view <number> --repo {owner}/{repo} --json title,body,headRefOid,author,files`, `gh pr diff`, and a fully parameterized `gh api -X POST .../pulls/{pull_number}/comments` call). It falls short of 5 because the instructed `gh api` command is not covered by the frontmatter allowed-tools (only `gh pr view/diff/comment`, `gh --version`, `gh auth status`), so the core posting step as written would be blocked, and placeholder substitution rules ({owner}/{repo}) are left implicit in places.

4 / 5

Workflow Clarity

The sequence is explicit and ordered — verify gh installed (stop if not), verify auth (wait, retry), resolve PR number/URL, view PR JSON, get diff, analyze, then post Part A inline comments and Part B summary — with real validation checkpoints on the prerequisites. Not 5: there is no verification step after posting (e.g., confirming comments were created successfully) and no checkpoint for the batch of inline comments before the summary goes up. Comfortably above 3 because prerequisite validation with a feedback loop (retry after login) is explicitly present.

4 / 5

Progressive Disclosure

The body is under 50 lines, has no bundle files (references/, scripts/, assets/ are absent), and needs none: everything fits in well-organized sections (Prerequisites, numbered review steps, Part A/Part B posting format, IMPORTANT constraints). Per the rubric's guidance for simple skills under 50 lines with no need for external references, well-organized sections score 5; nothing here should be split out into separate files.

5 / 5

Total

17

/

20

Passed

Description

53%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.

The description is concise, in third person, and clearly states what the skill does, but it is minimal: it lacks any 'when to use' trigger guidance and omits most of the skill's concrete capabilities (diff analysis, inline file comments, summary comment). It reads as a terse one-liner rather than a complete, trigger-rich description.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user asks to review, critique, or leave feedback on a pull request, or mentions reviewing a PR.'

Expand the 'what' with the skill's main capabilities, e.g. 'Analyzes the PR diff and posts inline file comments and a summary comment back to the PR.'

Include common synonyms such as 'pull request', 'code review', and 'PR feedback' to strengthen trigger-term coverage.

DimensionReasoningScore

Specificity

The description names the domain ('GitHub PR') and 1-2 concrete actions ('Perform a review', 'leaving comments on the PR'), but stops there — it does not mention analyzing diffs, posting inline file comments, or a summary. This matches the anchor 'Names domain and 1-2 concrete actions, but not comprehensive'; it is above anchor 2 because the actions are not purely generic.

3 / 5

Completeness

The 'what' is clear (review a GitHub PR and leave comments), but there is no 'when' — no 'Use when...' clause or equivalent trigger guidance, which per judging guidelines caps this dimension at 3. Anchor 4 requires an explicit, even if imperfect, 'when' clause, which is absent.

3 / 5

Trigger Term Quality

It contains natural terms users would say ('review', 'GitHub PR', 'comments'), but misses common variations and synonyms such as 'pull request', 'code review', or 'PR feedback'. This fits anchor 3 ('Some relevant keywords but missing common variations or synonyms'), short of anchor 4's 'Good keyword coverage'.

3 / 5

Distinctiveness Conflict Risk

'Perform a review on a GitHub PR, leaving comments on the PR' carves a reasonably clear niche (PR review posted back to GitHub), distinguishing it from generic document or data skills, though it could still overlap with general code-review skills. This sits between anchor 3 and anchor 5, noticeably above the midpoint: mostly distinct with minor overlap risk.

4 / 5

Total

13

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

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

14

/

16

Passed

Repository
DataDog/dd-trace-dotnet
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.