CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

gsd-code-review

Review source files changed during a phase for bugs, security issues, and code quality problems

Quality

65%

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

Fix and improve this skill with Tessl

tessl review fix ./pi/package-skills/gsd-code-review/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

63%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 body is a competent thin-dispatch layer: it gives explicit fail-closed error handling, concrete CLI invocations, and a clear read-then-execute sequence. Its weaknesses are duplicated flag documentation, inlined reviewer-lane configuration detail, and a single placeholder reference with no actual bundle files.

Suggestions

Document the `--depth` and `--files` flags once (in <objective>) and have <context> reference them, removing the duplicated explanations.

Move the reviewer-lane roster, slug/alias mapping, and model-key details (the 'Outside a phase' section) into a reference file linked from the body, keeping only the dispatch command inline.

Resolve the `@__GSD_PI_ENGINE_ROOT__` placeholder or ship the referenced workflow file in the bundle so progressive disclosure points at a real, navigable file.

DimensionReasoningScore

Conciseness

The body is dense and operational, but the `--depth` and `--files` flags are documented twice (once in <objective>, again in <context>), and the reviewer-lane paragraphs ('Optional reviewer-lane flags (#4209)...' and the 'Outside a phase' section) pack extensive configuration detail into a thin dispatch layer that could be tightened or moved out. This matches anchor 3 ('mostly efficient but includes some unnecessary explanation or could be tightened') rather than anchor 4's 'minor instances'.

3 / 5

Actionability

Concrete, near-executable commands are given — e.g. 'pipe canonical repo-relative file paths (one per line) to stdin of `gsd_run review-lane dispatch-step --repo-root <root> --depth <quick|standard|deep> --base-sha <sha> --run-dir "$(mktemp -d)" --cap-id code-review --point execute:post --explicit <slugs> --raw`' plus explicit dispatch instructions ('call `gsd_dispatch` with `role` ... and `task` ...'). Minor gaps remain (the main workflow lives in the external referenced file, and several placeholders are unresolved), so it falls just short of anchor 5's copy-paste-ready coverage and clearly above anchor 3's pseudocode level.

4 / 5

Workflow Clarity

The sequence is clear — 'Before doing anything else, read every file listed under <execution_context>', then 'Execute end-to-end' — with an explicit fail-closed checkpoint ('If `gsd_dispatch` reports the role unresolved or the child failed, STOP that step and surface the error') and defined optional/required escalation. This matches anchor 4 ('clear sequence with most checkpoints present; minor validation gaps'): the --fix auto-apply loop's validation is delegated to the external workflow rather than stated here, which keeps it below anchor 5.

4 / 5

Progressive Disclosure

There is a single clearly signaled one-level reference ('@__GSD_PI_ENGINE_ROOT__/gsd-core/workflows/code-review.md'), but no bundle files exist alongside the skill, the sole reference is an unresolved template placeholder, and substantial lane-dispatch/configuration detail is inlined that would fit better in a reference file. This matches anchor 3 ('some structure... content that should be separate is inline') better than anchor 4's 'most content is appropriately placed'.

3 / 5

Total

14

/

20

Passed

Description

50%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 states a clear 'what' with a moderately specific scope, but lacks any 'when to use' trigger guidance and misses natural trigger phrases and concrete outputs that would sharpen it. It sits squarely at the rubric midpoint across all dimensions.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user asks to review the code changed in a phase, run a code review, or asks for bugs/security findings before finishing a phase.'

Mention concrete outputs and capabilities already documented in the body — severity-classified findings written to a REVIEW.md artifact and quick/standard/deep depth levels — to lift specificity.

Include natural synonyms users would say ('code review', 'review my changes', 'security review') alongside 'bugs' and 'security issues' to improve trigger term coverage.

DimensionReasoningScore

Specificity

The description names the domain and a small set of actions — 'Review source files changed during a phase for bugs, security issues, and code quality problems' — but these are review categories rather than several distinct concrete actions (no mention of the REVIEW.md artifact, severity classification, or depth levels that the body documents). This matches anchor 3 ('names domain and 1-2 concrete actions, but not comprehensive') better than anchor 4, which expects a list of several specific actions with only minor gaps.

3 / 5

Completeness

The 'what' is clearly stated (review changed source files for bugs, security issues, quality), but there is no 'Use when...' clause or any equivalent trigger guidance for when to invoke the skill. Per the rubric guidelines, a missing 'Use when' clause caps completeness at 3.

3 / 5

Trigger Term Quality

Relevant keywords like 'review', 'source files', 'bugs', and 'security issues' are present, but common variations a user would naturally say ('code review', 'PR', 'diff', 'security review') are missing, and 'phase' is GSD-specific jargon rather than a natural trigger term. This fits anchor 3 ('some relevant keywords but missing common variations or synonyms'), not anchor 4's 'good keyword coverage'.

3 / 5

Distinctiveness Conflict Risk

The 'changed during a phase' scoping gives it a distinct niche versus generic code-review skills, but the phrasing still reads as a general code review ('bugs, security issues, and code quality problems') that overlaps with many similar review skills. Anchor 3 ('somewhat specific but could still overlap with similar skills') fits better than anchor 4's 'mostly distinct'.

3 / 5

Total

12

/

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

open-gsd/gsd-core

Path

pi/package-skills/gsd-code-review/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.