CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

review-change

Review a proposed code change and its evidence when asked to assess implementation quality.

Quality

72%

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 ./internal/adaptertest/testdata/all-agents/package/skills/review-change/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

72%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 admirably lean and points to a real script with a clear output contract, but the workflow is effectively truncated: a promised multi-step process ends after one step with no failure handling, and the existing REFERENCE.md is orphaned. It works as a stub, not as a complete skill.

Suggestions

Resolve the workflow inconsistency: either make this explicitly a single-step skill (drop 'Process steps in order') or add the missing steps that 'Finish here.' truncates.

Add a failure branch, e.g., 'If the check exits non-zero, report the failure and stop — do not proceed.'

Reference references/REFERENCE.md from the body (e.g., 'Evidence standards: see [REFERENCE.md](references/REFERENCE.md)') or remove it from the bundle.

DimensionReasoningScore

Conciseness

The body is roughly six lines with zero padding: 'Process steps in order. Do not skip ahead.' plus one step and its output contract. Every token earns its place and it assumes Claude's competence.

5 / 5

Actionability

'Run `skills/review-change/scripts/check.sh`' with the stated contract ('It emits `{"ok":true}` on stdout and exits non-zero on failure') is concrete and executable, but the failure path is undefined and the path is layout-dependent — mostly executable with minor gaps.

4 / 5

Workflow Clarity

'Process steps in order. Do not skip ahead.' implies a multi-step process, yet only 'Step 1 — Check the Change' exists before 'Finish here.', and no checkpoint says what to do when the script exits non-zero. Sequence is present but checkpoints are missing or implicit.

3 / 5

Progressive Disclosure

The body signals the script correctly, but the bundle's references/REFERENCE.md is never mentioned in the body — a provided reference that is not clearly signaled. Scoring against the actual bundle structure, this fits 'references present but not clearly signaled'.

3 / 5

Total

15

/

20

Passed

Description

57%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 covers what and when adequately and avoids fluff, but its action vocabulary is thin and trigger coverage lacks natural variations. It reads as a one-line domain statement rather than a capability inventory.

Suggestions

Enumerate concrete actions (e.g., 'check diffs, verify tests pass, audit evidence artifacts') instead of the single generic verb 'Review'.

Add natural trigger terms users would say, such as 'diff', 'PR', 'changeset', or 'implementation quality review'.

Tighten the when-clause to a 'Use when...' pattern with specific situations (e.g., 'Use when reviewing a diff, a pull request, or evidence for a proposed change').

DimensionReasoningScore

Specificity

"Review a proposed code change and its evidence" names the domain but the actions ('Review', 'assess') are generic — no concrete sub-actions like checking diffs or running tests, matching the 'Names the domain but actions are minimal or generic' anchor.

2 / 5

Completeness

Both 'what' ('Review a proposed code change and its evidence') and 'when' ('when asked to assess implementation quality') are present, but the when-clause could be more explicit and cover concrete trigger phrases.

4 / 5

Trigger Term Quality

Relevant keywords exist ('review', 'code change', 'assess implementation quality') but common variations users would naturally say ('diff', 'PR', 'code review', file extensions) are missing, fitting 'Some relevant keywords but missing common variations or synonyms'.

3 / 5

Distinctiveness Conflict Risk

The focus on reviewing proposed changes with evidence carves a mostly distinct niche, with only minor overlap risk against general code-review or quality-analysis skills.

4 / 5

Total

13

/

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

jbaruch/agentic-context-registry

Path

internal/adaptertest/testdata/all-agents/package/skills/review-change/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.