skills.rest
petekp
petekpCOMMUNITY
Explore all skills by this author →
Website𝕏@petekpGitHub

fix

Run Circuit's fix workflow to reproduce, patch, and verify bugs.

Updated Apr 18, 2026
Search Tags:#bug fixing#agent orchestration#flaky tests#regression#test verification#crash remediation
One-click install
npx skills add https://github.com/petekp/circuit-next --skill fix-petekp
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: fix
Source: https://github.com/petekp/circuit-next/tree/main/plugins/circuit/skills/fix
Command: npx skills add https://github.com/petekp/circuit-next --skill fix-petekp

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill helps you resolve bugs, regressions, failing tests, crashes, and flaky or broken behavior by guiding Circuit through reproduction, isolation, targeted change, verification, and reporting.

Core Features & Use Cases

  • Reproduce and isolate: Determines what’s failing and why before making changes.
  • Make the smallest safe patch: Relays a focused fix to a worker and then verifies it.
  • Mode-driven verification and review: Runs the full Fix flow by default, with Lite/Deep/Autonomous options for different depth and autonomy needs.
  • Use case: A production issue is reported and you need the minimal code change plus evidence that tests/verification passed, including a review step when required.

Quick Start

Ask for a bug fix by telling Circuit: fix the failing login flow that sometimes throws a token error.

Frequently Asked Questions about fix

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I fix failing tests and verify the regression is resolved?▼

To fix failing tests and verify regressions, this Skill runs a dedicated workflow that reproduces the failure, isolates the root cause, applies a minimal safe patch, and then runs verification to provide evidence the tests pass.

What is the best way to get a minimal safe code change for a production crash?▼

The best way to get a minimal safe code change for a production crash is using a fix workflow that isolates the issue, applies a targeted patch via a worker, and surfaces checkpoint and completion evidence verifying the broken behavior is resolved.

Can I use autonomous modes to remediate broken production behavior without manual review?▼

Yes, you can use the autonomous rigor mode to remediate broken production behavior, which streams progress and handles the fix flow end-to-end, though a review step is enforced when required to ensure safety.

Why does the bug fixing workflow require safe shell quoting for the goal?▼

The bug fixing workflow requires safe shell quoting for the goal to ensure the fix flow executes securely without injection risks, properly relaying the focused fix to the worker before running test verification and evidence generation.

Do I need to reproduce a test failure before applying a regression fix?▼

Yes, you need to reproduce a test failure before applying a regression fix because the workflow determines what is failing and why, ensuring the smallest safe patch is verified against the actual failure rather than applying a blind change.

Related Skills

Explore More Software Engineering Skills→
JKHeadleyJKHeadley
75

systematic-debugging

Diagnose bugs through a structured root-cause analysis workflow.

Community
Advanced
haidonglethqbhaidonglethqb

code-review

Define code-review scope and trigger verification steps across impacted files.

Community
Advanced
KYRIE66nbKYRIE66nb

ultraqa

Run automated QA cycles with testing, linting, and fixes until quality goals are met.

Community
Advanced
SnakeOSnakeO
9

debug:django

Diagnose Django errors with a four-phase debugging workflow.

Community
Intermediate