CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

differential-review

Performs security-focused differential review of code changes (PRs, commits, diffs). Adapts analysis depth to codebase size, uses git history for context, calculates blast radius, checks test coverage, and generates comprehensive markdown reports. Automatically detects and prevents security regressions.

Quality

73%

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 ./.claude/cloud-profile/skills/differential-review/SKILL.md

The canonical home for this skill is differential-review in trailofbits/skills

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.

A well-organized, token-efficient overview with strong navigation design and honest coverage guidance, undermined by a broken bundle: all four referenced detail files are missing, so the progressive-disclosure structure is non-functional and the actionable depth it promises is unavailable.

Suggestions

Ship the four referenced files (methodology.md, adversarial.md, reporting.md, patterns.md) in references/ so the Decision Tree and Supporting Documentation links resolve, or inline the essential per-phase steps into SKILL.md and drop the broken links.

Add the concrete commands for the core phases into whichever file survives — e.g. the actual git log/blame commands and the quantitative blast-radius calculation — so guidance is executable without inference.

Trim 'Example Usage' and 'Tips for Best Results', which substantially repeat the Decision Tree, Quick Reference, and Quality Checklist, to improve token efficiency.

DimensionReasoningScore

Conciseness

The body is dense and efficient — tables for rationalizations, size strategy, and risk triggers with no exposition of concepts Claude already knows. It falls short of 5 because sections like 'Example Usage' and 'Tips for Best Results' restate guidance already covered by the Decision Tree and Quality Checklist, and could be trimmed.

4 / 5

Actionability

Some guidance is concrete (risk triggers, the issue-writer command, red-flag escalation criteria), but the executable detail for the core phases — how to run git history analysis, calculate blast radius quantitatively, or structure the report — is deferred to methodology.md, adversarial.md, reporting.md, and patterns.md, none of which exist in the bundle. This leaves key execution details missing, matching the 'incomplete / missing key details' anchor.

3 / 5

Workflow Clarity

A clear sequence is laid out (Pre-Analysis through Phase 6) with a workflow diagram, a decision tree for navigation, and a Quality Checklist acting as a pre-delivery checkpoint. It is not a 5 because the per-phase steps and explicit validation feedback loops live in the referenced files rather than being present, leaving minor checkpoint gaps.

4 / 5

Progressive Disclosure

The structure is good on paper — a decision tree and a 'Supporting Documentation' section signal one-level-deep references to four files — but scoring against the actual bundle, none of the four referenced files (methodology.md, adversarial.md, reporting.md, patterns.md) exist in references/ or anywhere else. The broken links mean the disclosure structure does not function, which is more than the 'minor organization gaps' of a 4.

3 / 5

Total

14

/

20

Passed

Description

71%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.

A strong, third-person description with concrete, comprehensive capability statements and good natural trigger terms. Its main deficiency is the complete absence of explicit 'when to use' guidance, which caps completeness and slightly weakens its usefulness for skill routing.

Suggestions

Add an explicit trigger clause, e.g. 'Use when the user asks to review a PR, commit, or diff, especially for security issues, or mentions blast radius or security regressions.'

Include common synonyms such as 'pull request', 'code review', and 'audit' alongside 'PRs, commits, diffs' to broaden natural trigger coverage.

Clarify the boundary with generic code-review skills (e.g. 'for security-critical changes rather than general code review') to reduce routing conflicts.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — 'uses git history for context, calculates blast radius, checks test coverage, and generates comprehensive markdown reports' — plus a clear domain ('security-focused differential review of code changes (PRs, commits, diffs)'). Coverage is comprehensive with no significant gaps; it is not the 4 anchor because nothing meaningful is missing from the action list.

5 / 5

Completeness

The 'what' is clearly and concretely stated, but there is no 'Use when...' clause or equivalent explicit trigger guidance anywhere in the description — the 'when' is entirely absent, not merely weakly implied. Per the rubric guideline, a missing 'Use when...' clause caps completeness at 3.

3 / 5

Trigger Term Quality

Good natural keyword coverage: 'security', 'review', 'code changes', 'PRs', 'commits', 'diffs' — terms users would naturally say. It falls short of the 5 anchor because common synonyms and variants are missing, e.g. 'pull request', 'code review', 'audit', or 'vulnerability'.

4 / 5

Distinctiveness Conflict Risk

The 'security-focused differential review' framing carves a clear niche with distinct triggers (diffs, blast radius, security regressions), giving minimal conflict risk with unrelated skills. It is not a 5 because it overlaps with generic code-review skills, which also trigger on 'PRs', 'commits', and 'review of code changes'.

4 / 5

Total

16

/

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

relative_links

Relative link issues: 4 possibly missing (adversarial.md, methodology.md, patterns.md, reporting.md)

Warning

Total

15

/

16

Passed

Repository

BigSimmo/PsychSift

Path

.claude/cloud-profile/skills/differential-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.