CtrlK
BlogDocsLog inGet started
Tessl Logo

cerebro-regression-tests

Add focused regression coverage for Cerebro review findings, bugs, and security edge cases.

68

1.11x
Quality

52%

Does it follow best practices?

Impact

94%

1.11x

Average score across 3 eval scenarios

SecuritybySnyk

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./.factory/skills/cerebro-regression-tests/SKILL.md
SKILL.md
Quality
Evals
Security

Evaluation results

82%

-4%

URL Parsing Regression in Multi-Tenant API Gateway

Security boundary regression tests for URL/tenant parser

Criteria
Without context
With context

Table-driven URL tests

85%

71%

Table-driven tenant tests

21%

14%

Embedded userinfo boundary

100%

100%

Host port suffix boundary

100%

100%

Empty host boundary

100%

100%

Oversized host boundary

100%

100%

Auth-in-query rejection

100%

100%

Regression assertion direction

100%

100%

Verbose test output file

90%

80%

100%

Pagination Overflow in Data Export Pipeline

Smallest package-level regression test for pagination bug

Criteria
Without context
With context

Single focused package

100%

100%

Local fixtures

100%

100%

Bug boundary assertion

100%

100%

Test would fail without fix

100%

100%

Minimal production fix

100%

100%

Fix covered by test

100%

100%

Verbose cache-disabled test run

100%

100%

Size-limit boundary case

100%

100%

100%

33%

Graph Traversal Edge-Case Coverage

Table-driven graph edge case regression with run documentation

Criteria
Without context
With context

Table-driven graph tests

0%

100%

Cycle detection case

100%

100%

Zero out-edge node case

100%

100%

Error-mapping case

100%

100%

Focused run command documented

50%

100%

make verify documented

0%

100%

Tests assert regression boundary

100%

100%

Test output file present

100%

100%

Repository
writer/cerebro
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.