CtrlK
BlogDocsLog inGet started
Tessl Logo

code-review

Review PRs, MRs, and Gerrit changes with focus on security, maintainability, and architectural fit. Leverages github, gitlab, or gerrit skills based on repository context. Use when asked to review my code, check this PR, review a pull request, look at a merge request, review a patchset, or provide code review feedback.

Install with Tessl CLI

npx tessl i github:odyssey4me/agent-skills --skill code-review
What are skills?

96

1.42x

Does it follow best practices?

Evaluation98%

1.42x

Agent success when using this skill

Validation for skill structure

SKILL.md
Review
Evals

Evaluation results

99%

52%

Set Up Review Tracking for a Microservices Repository

Repository context persistence

Criteria
Without context
With context

Config file path

0%

100%

Output file created

100%

80%

Top-level repositories key

0%

100%

Remote URL as repo key

0%

100%

references list present

0%

100%

Reference URL in references

42%

100%

standards list present

62%

100%

JWT standard in standards

100%

100%

Parameterized queries standard

100%

100%

notes list present

62%

100%

Migration note in notes

100%

100%

YAML format

100%

100%

Without context: $2.6225 · 8m 24s · 87 turns · 452 in / 24,297 out tokens

With context: $0.3742 · 1m 18s · 20 turns · 55 in / 4,223 out tokens

95%

55%

Review a Code Change for the Acme Auth Service

CI-aware review format

Criteria
Without context
With context

review.md exists

100%

100%

Review heading format

50%

100%

Summary section

60%

100%

CI Status section

20%

100%

Failed tests noted

85%

100%

Findings section with subsections

0%

100%

Hardcoded secret flagged

100%

100%

Return type bug not re-flagged as new issue

25%

58%

Style issue not flagged

0%

100%

Verdict section

0%

100%

File:line references

20%

100%

Without context: $0.1687 · 1m 8s · 10 turns · 15 in / 3,108 out tokens

With context: $0.3137 · 1m 27s · 12 turns · 1,192 in / 4,103 out tokens

100%

17%

Review a Code Change for the TechPayments Transaction Service

Security-first review priorities

Criteria
Without context
With context

review.md exists

100%

100%

Security section before others

60%

100%

SQL injection flagged

100%

100%

Data exposure flagged

100%

100%

SSRF flagged

100%

100%

Missing authorization flagged

100%

100%

Import/whitespace not flagged

0%

100%

File:line references

66%

100%

CI Status section

60%

100%

Verdict present

100%

100%

Security verdict for critical issues

100%

100%

Without context: $0.1955 · 1m 28s · 9 turns · 14 in / 4,154 out tokens

With context: $0.3787 · 1m 41s · 16 turns · 988 in / 4,636 out tokens

94%

19%

Reviewing a PR Against Team Engineering Standards

Saved context applied in review

Criteria
Without context
With context

Context loaded

100%

100%

Parameterized query standard violation flagged

100%

100%

TokenValidator standard violation flagged

100%

100%

Sensitive field logging standard violation flagged

100%

100%

Hardcoded secret flagged

100%

100%

File:line references

0%

25%

Correct review format

40%

100%

Security section present

0%

100%

Verdict present

57%

100%

REQUEST_CHANGES verdict

62%

100%

Without context: $0.2000 · 1m 9s · 8 turns · 13 in / 3,684 out tokens

With context: $0.3235 · 1m 31s · 14 turns · 16 in / 4,240 out tokens

100%

Cleaning Up Stale Review Context

Forget and show context commands

Criteria
Without context
With context

Updated config file created

100%

100%

Context display file created

100%

100%

gRPC note removed

100%

100%

Legacy reference removed

100%

100%

Remaining references preserved

100%

100%

Standards untouched

100%

100%

Payments note preserved

100%

100%

Other repo unchanged

100%

100%

Top-level YAML structure

100%

100%

Context display shows remaining items

100%

100%

Without context: $0.1124 · 32s · 10 turns · 14 in / 1,711 out tokens

With context: $0.4021 · 1m 28s · 21 turns · 59 in / 4,242 out tokens

100%

27%

Review a PR for a Notification Service Refactor

Maintainability and architecture review

Criteria
Without context
With context

Pending CI noted

37%

100%

Hardcoded credentials flagged

100%

100%

SSRF flagged

100%

100%

Excessive complexity flagged

100%

100%

Poor naming flagged

100%

100%

Mixed concerns flagged

100%

100%

Dead code flagged

100%

100%

No style/whitespace findings

75%

100%

Correct review format

25%

100%

Findings sections present

33%

100%

Verdict present

40%

100%

File:line references

0%

100%

Without context: $0.2976 · 2m 6s · 10 turns · 15 in / 6,111 out tokens

With context: $0.3868 · 1m 51s · 16 turns · 315 in / 5,104 out tokens

Evaluated
Agent
Claude Code
Model
Unknown

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.