CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

review-change

Review a change.

Quality

40%

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/preserve/testdata/markdown-existing/package/skills/review-change/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

33%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 SKILL.md body is an empty stub — a single heading with no guidance, commands, examples, workflow, or supporting files. While it wastes zero tokens, it provides zero actionable content, so the skill cannot function as written. The what/when must be authored, not inferred.

Suggestions

Add the actual review procedure as concrete, executable guidance — e.g. the commands to obtain the diff ("git diff", "gh pr diff <n>"), what to inspect, and how to report findings.

Add a sequenced workflow with a validation checkpoint (e.g. re-run the diff after fixes, confirm findings against the actual code), which is essential since reviewing changes is a multi-step, error-prone process.

Either write enough body content to make the skill self-contained or split detail into a references/ file and link it clearly from the body — currently there is nothing to navigate to.

DimensionReasoningScore

Conciseness

The body is a single heading ("# Review change") with zero padding and no explanation of concepts Claude already knows — nothing to trim. It matches anchor 5 ("Lean and efficient; every token earns its place") on the token-waste axis, though only because there is almost no content.

5 / 5

Actionability

The body contains no code, no commands, no steps, and no instructions of any kind — only a title. This matches anchor 1 ("Entirely vague or abstract; no concrete code or commands; only describes rather than instructs") in its most extreme form: it doesn't even describe the task.

1 / 5

Workflow Clarity

No steps, no sequence, and no validation are present — the body is empty of any process guidance, matching anchor 1 ("Steps missing or incoherent; no sequence; no validation for risky operations"). The simple-skill exception does not apply because even the single action is unstated, not merely unambiguous.

1 / 5

Progressive Disclosure

There is minimal structure: one bare heading, no sections, no overview, and no referenced files (the references/, scripts/, assets/ directories do not exist). This matches anchor 2 ("Minimal structure") rather than anchor 1, since there is no monolithic wall of text or nested-reference chain to navigate.

2 / 5

Total

9

/

20

Passed

Description

20%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 near-empty placeholder: one generic verb and an unspecified object with no concrete capabilities, no trigger conditions, and no natural keywords beyond "review" and "change". It closely resembles the rubric's bad overall examples ("Processes data", "Helps with documents").

Suggestions

State concrete capabilities, e.g. "Reviews a code change (diff, staged changes, or PR) for correctness bugs, style issues, and missing tests."

Add an explicit trigger clause, e.g. "Use when the user asks to review a diff, a pull request, staged or uncommitted changes, or says 'review this change'."

Include the natural terms users would actually say — diff, PR, pull request, code review, staged changes — to improve trigger matching and distinctiveness.

DimensionReasoningScore

Specificity

The entire description is "Review a change." — a single generic verb applied to an unspecified object, no more concrete than the canonical bad example "Processes data". It does not name a concrete domain (unlike anchor 2's "Processes PDF files"), so it sits at the vague end of the scale.

1 / 5

Completeness

There is only a vague "what" ("Review a change.") and no "when" clause at all, exactly matching anchor 2 ("Has a vague 'what' and no 'when'"). The missing 'Use when...' guidance caps this dimension at 3, and the vague 'what' pushes it below that.

2 / 5

Trigger Term Quality

The only keywords are "review" and "change" — one or two generic terms, missing the natural phrases users would say (diff, PR, pull request, code review, staged changes, commit). This matches anchor 2 ("One or two generic keywords; missing the natural phrases users say").

2 / 5

Distinctiveness Conflict Risk

"Review a change" is very broad and would overlap with many similar skills (code review, security review, simplify, PR review), matching anchor 2 ("Very broad; high overlap risk with many similar skills"). It is not anchor 1 because it is constrained to reviewing changes rather than conflicting with virtually any skill.

2 / 5

Total

7

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

description_field

'description' is very short (16 chars), consider making it more detailed

Warning

Total

15

/

16

Passed

Repository

jbaruch/agentic-context-registry

Path

internal/preserve/testdata/markdown-existing/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.