code-reviewer

Code review automation for TypeScript, JavaScript, Python, Go, Swift, Kotlin. Analyzes PRs for complexity and risk, checks code quality for SOLID violations and code smells, generates review reports. Use when reviewing pull requests, analyzing code quality, identifying issues, generating review checklists.

Install with Tessl CLI

npx tessl i github:alirezarezvani/claude-skills --skill code-reviewer

What are skills?

1.78x

Quality

93%

Does it follow best practices?

Impact

84%

1.78x

Average score across 6 eval scenarios

SKILL.md

Review

Evals

Evaluation results

100%

90%

Authentication Service Security Review

PR security risk analysis with JSON output

Criteria

Without context

With context

JSON output format

100%

Complexity score present

100%

Risk severity categories

100%

File review order

100%

Hardcoded secret detected

100%

SQL injection detected

100%

TypeScript any type detected

100%

Console statements detected

100%

ESLint suppression detected

100%

TODO/FIXME comments detected

100%

Without context: $0.9085 · 4m 18s · 46 turns · 411 in / 11,282 out tokens

With context: $0.5229 · 1m 55s · 24 turns · 310 in / 4,214 out tokens

95%

25%

Order Processing Module Quality Assessment

Code quality structural smell detection

Criteria

Without context

With context

JSON output format

100%

Total smells reported

Quality score present

50%

100%

God class detected

100%

Long function detected

33%

100%

Too many parameters detected

100%

High complexity detected

100%

Magic numbers detected

100%

God class threshold in message

100%

Parameter count in smell message

100%

Without context: $0.4816 · 2m 27s · 21 turns · 24 in / 6,628 out tokens

With context: $0.4477 · 1m 14s · 16 turns · 7,102 in / 3,693 out tokens

100%

31%

Reporting Pipeline PR — Formal Review Report

Full review report with verdict and pre-computed analyses

Criteria

Without context

With context

Markdown report created

100%

PR analysis JSON created

60%

100%

Quality analysis JSON created

80%

100%

Verdict present

40%

100%

Score present

100%

Verdict matches score

20%

100%

Action items section

100%

Issue summary section

60%

100%

PR statistics incorporated

100%

Review order present

40%

100%

Without context: $0.5328 · 3m 1s · 26 turns · 26 in / 9,649 out tokens

With context: $0.8467 · 1m 55s · 28 turns · 6,689 in / 5,509 out tokens

24%

-16%

Data Pipeline Quality Assessment

Language-scoped deep nesting and large file detection

Criteria

Without context

With context

Language flag used

JSON flag used

100%

Output file created

100%

Large file detected

Deep nesting detected

100%

Missing error handling detected

Unused imports detected

100%

Magic numbers detected

100%

Nesting threshold in message

62%

File line count in message

Without context: $0.8187 · 4m 5s · 27 turns · 27 in / 14,888 out tokens

With context: $0.5952 · 1m 42s · 23 turns · 276 in / 4,257 out tokens

100%

56%

Feature Branch Risk Review

Branch-specific PR analysis with commit validation

Criteria

Without context

With context

Base/head flags used

100%

JSON flag used

100%

Output file created

100%

Complexity score present

100%

Risk categories present

50%

100%

Hardcoded credentials detected

66%

100%

SQL injection detected

80%

100%

Commit message validation present

41%

100%

File review order present

100%

Repository path used

100%

Without context: $0.4908 · 2m 36s · 25 turns · 28 in / 8,178 out tokens

With context: $0.6902 · 2m 2s · 35 turns · 38 in / 5,786 out tokens

88%

34%

Notification Service Code Review

TypeScript async antipatterns and language standards review

Criteria

Without context

With context

Report generator used

100%

No pre-computed flags

Markdown format flag used

100%

Output file flag used

100%

Markdown report created

100%

Verdict present

100%

Score present

100%

Floating promise issue reported

100%

TypeScript any type reported

100%

Action items section

100%

Review order present

100%

Without context: $0.6554 · 3m 28s · 33 turns · 29 in / 10,436 out tokens

With context: $1.5610 · 5m 1s · 42 turns · 6,705 in / 16,609 out tokens

Evaluated: 1 day ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Authentication Service Security Review Order Processing Module Quality Assessment Reporting Pipeline PR — Formal Review Report Data Pipeline Quality Assessment Feature Branch Risk Review Notification Service Code Review

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.