CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

reviewer

Load when reviewing a diff, classifying findings, or returning a verdict.

Quality

67%

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/reviewer/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

71%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 dense, well-structured operational contract: executable commands with explicit validation loops, precise artifact-path rules, and disciplined delegation of detail to reference files. Its main weaknesses are broken navigation — several linked files (workflows/, schemas/review-finding.md) are missing from the bundle while two reference files are orphaned — and a reliance on ecosystem jargon and sibling-skill anchors that the bundle alone does not fully support.

Suggestions

Ship the referenced files (workflows/review.md, workflows/mutation-stability.md, schemas/review-finding.md) in the bundle, or restructure so every path the body links to resolves within it.

Reference references/perf-qa.md and references/ui-screenshots.md from SKILL.md (or from a bundled workflow file) so no bundle file is orphaned and each is discoverable in one hop.

DimensionReasoningScore

Conciseness

The body is dense with rules and assumes Claude's competence throughout — no padding, no explanation of known concepts, e.g. "Verify before reporting: if the repo contains the caller, config, test, or doc that settles a suspicion, read it." Not 5 because the hooks paragraph ("Where installed and running, `reviewer-read-only` refuses edits…") and the repeated "judge against code-quality, not the change" theme across several Ethos bullets could be consolidated and trimmed.

4 / 5

Actionability

Concrete, copy-paste-ready commands with explicit success criteria: `orch-runtime review-artifact --text check --file [ARTIFACT_PATH] --worktree [WORKTREE_PATH]` with "Fix until this prints `\"ok\": true`", and `review-depth --base REF --head REF --repo DIR`. Not 5 because roughly half the body (the Ethos rules) is policy prose with no worked example of a finding or artifact, leaving minor gaps between rule and execution.

4 / 5

Workflow Clarity

The Output Contract is a clear sequence (capture starting fields → write artifact → self-validate → return) with an explicit validate-fix-retry checkpoint and a precise artifact path format. Not 5 because the core review workflow (diff → findings → verdict) is delegated to `workflows/review.md`, which is not present in the bundle, so the end-to-end sequence with checkpoints is never shown in one place.

4 / 5

Progressive Disclosure

In-document structure is good and one-level references to real bundle files exist (references/finding-disposition.md, scripts/review-depth, scripts/mutation-stability), but the body links to workflows/review.md, workflows/mutation-stability.md and schemas/review-finding.md that are absent from the bundle, and references/perf-qa.md and ui-screenshots.md are unreachable from SKILL.md (reachable only via a missing workflows/ file). Scored against the actual bundle structure, navigation is broken in both directions, matching anchor 3 rather than 4.

3 / 5

Total

15

/

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 is a pure load-trigger: it states when to load the skill with three moderately specific activities but never states what the skill does, and it misses the domain's natural trigger vocabulary (code review, PR, pull request). It is clear and honest, just thin — a reader cannot tell what capabilities the skill actually provides or how it differs from other review skills.

Suggestions

Add an explicit capability clause naming what the skill does, e.g. "Reviews code diffs, classifies and verifies findings into a JSON artifact, and returns a pass/block verdict per its finding schema. Load when reviewing a diff, classifying findings, or returning a verdict."

Include natural trigger synonyms users actually say — "code review", "PR review", "pull request" — alongside "reviewing a diff".

State what kind of reviewer this is (shared contract for review specialists) so it is distinguishable from other review skills and does not trigger for the wrong one.

DimensionReasoningScore

Specificity

"reviewing a diff, classifying findings, or returning a verdict" names the domain plus three activities, but they are high-level gerunds with no objects and omit the skill's real capabilities (artifact writing, QA re-review, mutation-stability probes). It sits at anchor 3 rather than 4 because the actions are more abstract than anchor 4's concrete operations and coverage is not minor-gap but substantial.

3 / 5

Completeness

The "when" is explicit ("Load when reviewing a diff, classifying findings, or returning a verdict") but the "what" is never stated — the skill's function is only implied by its trigger activities. This falls between anchor 2 (when-only, as in "Use when working with documents") and anchor 4 (both present): materially better than 2 because the triggers are specific, below 4 because no capability statement exists.

3 / 5

Trigger Term Quality

"reviewing a diff" is a phrase users naturally say, but common variations and synonyms — "code review", "PR", "pull request", "audit" — are missing, matching the anchor for some relevant keywords with missing variations. Not 4 because the natural vocabulary of the domain is largely absent beyond the single diff phrase.

3 / 5

Distinctiveness Conflict Risk

"diff", "findings", "verdict" are generic to any code-review skill; the description never says which kind of review (code, QA, docs, security), so overlap with sibling review skills in the same ecosystem is realistic. It is somewhat specific — clearly a review skill — but not "mostly distinct" as anchor 4 requires.

3 / 5

Total

12

/

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

metadata_version

'metadata.version' is missing

Warning

relative_links

Relative link issues: 6 possibly missing (./schemas/review-finding.md, ./workflows/review.md, schemas/review-finding.md#item-fields-blockerssuggestions, schemas/review-finding.md#measurement-claims, workflows/mutation-stability.md, +1 more), 11 suspicious

Warning

Total

14

/

16

Passed

Repository

vanillagreencom/kendex

Path

.agents/skills/reviewer/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.