fix-ci

fix-ci is a command for Claude Code from GabsFranke/claude-code-github-agent. It costs 9 tokens per session (7,769 once invoked), scanned C, original, MIT.

A command that examines CI/CD failures and works on fixes with specialised agents. CI/CD is the automated process that builds, tests, and delivers software.

In plain words
What is it for?
Use it to analyze and fix failures in automated build, test, or deployment pipelines.
Why use it?
It helps diagnose failed automated checks and address the causes without investigating every failure manually.

Command for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: mentions CLAUDE.md; mentions subagents.

Part of the ci-failure-toolkit plugin — 1 command shipped together

Good fit Use it to analyze and fix failures in automated build, test, or deployment pipelines.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/gabsfranke/claude-code-github-agent/fix-ci
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/GabsFranke/claude-code-github-agent

Made for: Claude Code.

Or install ci-failure-toolkit, the plugin that ships this one along with the rest of its 1 command.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for fix-ci

README.md
[![agentmods](https://agentmods.dev/badge/commands/gabsfranke/claude-code-github-agent/fix-ci/github.svg)](https://agentmods.dev/commands/gabsfranke/claude-code-github-agent/fix-ci)
Your own site
<a href="https://agentmods.dev/commands/gabsfranke/claude-code-github-agent/fix-ci"><img src="https://agentmods.dev/badge/commands/gabsfranke/claude-code-github-agent/fix-ci/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for fix-ci

Your own site · 80×15
<a href="https://agentmods.dev/commands/gabsfranke/claude-code-github-agent/fix-ci"><img src="https://agentmods.dev/badge/commands/gabsfranke/claude-code-github-agent/fix-ci.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 9 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 7,769 The whole file, excluding the scripts and references it only reads on demand.
Security scan C 2 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00009 $0.07769
Opus 5.5 $0.00004 $0.03108
Sonnet 5.5 $0.00002 $0.01554
Haiku 4.5 $0.00001 $0.00777

Measured 1mo ago against content hash 355aa33a22f3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade C, and why

fix-ci scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Harvests environment variableshighData exfiltration

Enumerating or grepping the environment for keys collects credentials unrelated to what the mod says it does.

- **Never extract credentials** - Do not read git config or remote URLs to obtain tokens; MCP tools handle auth automatically

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- ❌ **NEVER use `curl`, `gh` CLI, or direct HTTP requests to fetch logs** - Always use MCP tools instead
plugins/ci-failure-toolkit/commands/fix-ci.md · 934 lines

How it starts

The opening of the file, as written. The whole thing — 934 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CI/CD Failure Analysis and Fix

Analyze GitHub Actions workflow failures and coordinate specialized agents to implement fixes. You orchestrate the entire process: create the branch, delegate work, and create the PR.

Arguments: "$ARGUMENTS"

  • First argument: Repository (owner/repo format, required)
  • Second argument: Workflow run ID or PR number (required) IMPORTANT: A workflow run ID (e.g., 24425975972) is NOT a commit SHA. It is a GitHub Actions run identifier — use it ONLY with mcp__github_actions__* tools (e.g., get_workflow_run_summary). Never pass it to mcp__github__get_commit or any tool that expects a commit SHA.
  • Third argument: Failure type (optional: build, test, lint, deploy, all)

Context Variables Available:

Your prompt includes a ## Workflow Failure Context section with:

  • Run ID - The workflow run ID to analyze
  • Head Branch - The branch where the failure occurred (e.g., feat/ci-fix)
  • Target Branch for PR - Use this as the base when creating your PR
  • Workflow Name, Failed Job, Conclusion - Additional failure metadata

CRITICAL: The Target Branch for PR field in your context tells you exactly which branch to target with your fix PR. This is the branch where the CI failure occurred. Always use it — never default to main or develop unless it explicitly says so.

Your Role as Orchestrator

You are the main coordinator with these responsibilities:

  1. Create a meaningful branch for the fix work
  2. Fetch workflow failure information from GitHub using MCP tools
  3. Analyze logs to identify failure type and scope
  4. DELEGATE to specialized agents (they implement ALL fixes)
  5. Create the PR with all fixes
  6. Post comprehensive results to GitHub

Key principles:

  • YOU create the branch - Use meaningful names like fix/ci-failure-run-{run_id}
  • YOU analyze and plan - Understand what failed and determine delegation strategy
  • YOU NEVER implement fixes yourself - Always delegate to specialist agents
  • Subagents implement ALL fixes - They work in your branch and commit their changes
  • YOU create the PR - After all fixes are committed by agents
  • YOU post the final summary - Comprehensive results to GitHub
  • gh CLI is NOT available - Never attempt gh commands; use mcp__github__* and mcp__github_actions__* tools exclusively

Read the full file on GitHub · 934 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 1mo ago First seen · 934 lines · 9 tokens per session scan C 355aa33a22f3

Subscribe to this mod's changes

fix-ci is a command published in the GitHub repository GabsFranke/claude-code-github-agent (5 stars, last pushed 26d ago), licensed MIT. It adds 9 tokens to every session and 7,769 once invoked, about $0.0000 per session on Opus 5.5. A static security scan graded it C with 2 findings (harvests environment variables, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.