CtrlK
BlogDocsLog inGet started
Tessl Logo

code-review

Code review practices with technical rigor and verification gates. Use for receiving feedback, requesting code-reviewer subagent reviews, or preventing false completion claims in pull requests.

81

1.06x
Quality

72%

Does it follow best practices?

Impact

94%

1.06x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./plugins/code-review/skills/code-review/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

91%

8%

Responding to an Automated Code Review

Code review reception protocol

Criteria
Without context
With context

No performative agreement

50%

33%

Stops on unclear items

66%

100%

No partial implementation while unclear

100%

100%

YAGNI check on SessionCacheManager

80%

100%

YAGNI check on BaseCache

100%

100%

Technical pushback on at least one item

100%

100%

No defensive tone in pushback

87%

100%

Implements unambiguous valid items

100%

100%

Technical acknowledgment on implemented items

70%

90%

External reviewer skepticism

90%

100%

93%

-4%

Implement and Verify an Email Validation Utility

Verification gates before completion claims

Criteria
Without context
With context

Test command cited

100%

100%

Test output included

100%

100%

Existing tests verified

100%

100%

Build/typecheck command cited

100%

100%

Per-criterion checklist

100%

100%

No unverified success claims

100%

100%

Bun preferred for tests

100%

100%

Evidence before claim pattern

70%

30%

No agent report trusted blindly

100%

100%

Gaps reported honestly

100%

100%

100%

15%

Feature Complete: Document the Code Review Request Process

Code-reviewer subagent request process

Criteria
Without context
With context

BASE_SHA command

100%

100%

HEAD_SHA command

100%

100%

Task tool dispatch mentioned

100%

100%

code-reviewer agent type

0%

100%

WHAT_WAS_IMPLEMENTED field

87%

100%

PLAN_OR_REQUIREMENTS field

62%

100%

DESCRIPTION field

83%

100%

Critical: fix immediately

100%

100%

Important: fix before proceeding

100%

100%

Minor: note for later

100%

100%

Push-back criteria mentioned

100%

100%

No 'skip because simple' shortcut

100%

100%

Repository
secondsky/claude-skills
Evaluated
Agent
Claude Code
Model
Claude Sonnet 4.6

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.