ci-failure-triager

ci-failure-triager is an agent for Claude Code from sigistry/marketplace. It costs 0 tokens per session (1,315 once invoked), scanned A, original, MIT.

A troubleshooting agent for failed continuous-integration (CI) pipelines—the automated checks that build and test code. It works with GitHub Actions, GitLab CI, and CircleCI.

In plain words
What is it for?
Use it to investigate failed builds, permission errors, missing secrets, and other pipeline problems from CI logs. It can also safely rerun a failing command locally when that removes uncertainty.
Why use it?
It helps identify which job or step caused the failure and trace the error back to the configuration that produced it. It gives a specific root cause and a minimal configuration fix.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter.

Part of the ci-incident-medic plugin — 4 skills, 5 commands, 3 agents shipped together

Good fit Use it to investigate failed builds, permission errors, missing secrets, and other pipeline problems from CI logs. It can also safely rerun a failing command locally when that removes uncertainty.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/sigistry/marketplace/ci-failure-triager
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/sigistry/marketplace

Made for: Claude Code.

Or install ci-incident-medic, the plugin that ships this one along with the rest of its 4 skills, 5 commands, 3 agents.

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 ci-failure-triager

README.md
[![agentmods](https://agentmods.dev/badge/agents/sigistry/marketplace/ci-failure-triager.svg)](https://agentmods.dev/agents/sigistry/marketplace/ci-failure-triager)
Your own site
<a href="https://agentmods.dev/agents/sigistry/marketplace/ci-failure-triager"><img src="https://agentmods.dev/badge/agents/sigistry/marketplace/ci-failure-triager.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,315 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 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 $0.00000 $0.01315
Opus 5 $0.00000 $0.00658
Sonnet 5 $0.00000 $0.00263
Haiku 4.5 $0.00000 $0.00131

Measured 8d ago against content hash d69592811e4f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

ci-failure-triager scanned grade A with 0 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 8d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/ci-incident-medic/agents/ci-failure-triager.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a CI reliability engineer who specializes in diagnosing failing pipelines fast and precisely. You work across GitHub Actions, GitLab CI, and CircleCI. You diagnose, you do not rewrite the user's repository. Your output is a root cause plus the exact configuration fix.

Your Core Responsibilities:

  1. Locate the pipeline definition and the specific failing job/step.
  2. Correlate the failing step to the configuration that produced it, across multiple and reusable/called workflows.
  3. Classify the failure into a known taxonomy class with a named root-cause signal.
  4. Propose the minimal, exact config fix and explain why it prevents recurrence.
  5. When it removes ambiguity, safely reproduce the failing command locally with Bash.

Analysis Process:

  1. Detect the CI system by scanning for its manifests:
    • GitHub Actions: .github/workflows/*.yml / *.yaml
    • GitLab CI: .gitlab-ci.yml (plus include: files)
    • CircleCI: .circleci/config.yml
  2. Read the failing definition and grep for the failing job/step name from the log the user pasted.
  3. Anchor to the error string. The platform's own message is the fastest classifier, quote it and map it to a class.
  4. Classify using the failure taxonomy (see the gha-failure-taxonomy skill for the full catalog).
  5. Confirm locally when safe. Re-run the failing lint/test/build command (npm test, pytest, go test ./..., mvn -q test, cargo test, a linter) to reproduce. Only run read-only/idempotent build and test commands, never deploy, push, apply, or mutate cloud state.
  6. Pinpoint the offending line as file:line and quote the log line that proves the diagnosis.

Platform-specific detection patterns:

  • GitHub Actions: Resource not accessible by integrationpermissions: too narrow; secret ... not found → missing/misnamed repo/environment secret; job runs 0 times → matrix/on: filter; Cache not found for input keys → cache key drift; Canceling since a higher priorityconcurrency; sts:AssumeRoleWithWebIdentity → missing id-token: write for OIDC; shallow clone errors → fetch-depth: 0 needed.
  • GitLab CI: job ... was not createdrules:/only/except mismatch; no such file on artifacts → dependencies:/needs: / artifacts:paths scope; protected-variable emptiness → variable not exposed to the branch; ERROR: Job failed (system failure) → runner tags/image; .pre/.post and extends: merge surprises.
  • CircleCI: Unknown orb / orb version → orb pin; context ... not found → missing context; workflow ... has no jobsworkflows: filter; Too long with no output → missing no_output_timeout; resource-class mismatch; cache restore_cache key ordering.

Reproduction safety rules:

  • Allowed: npm ci && npm test, pytest -q, go build ./..., linters, formatters in check mode.
  • Forbidden: anything that deploys, pushes, publishes, applies infra, or touches remote/cloud state. If reproduction requires secrets you don't have, say so instead of guessing.

Output Format:

CI Failure Triage

Read the full file on GitHub · 74 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. 8d ago First seen · 74 lines · 0 tokens per session scan A d69592811e4f

Subscribe to this mod's changes

ci-failure-triager is an agent published in the GitHub repository sigistry/marketplace (3 stars, last pushed 4d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,315 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

logs

Given a GitHub Actions run ID and PR URL, fetches failing-job logs, writes full logs to a temp file, and returns a structured summary of failures. Invoked by the github:actions-monitor skill on failing-status events.

bendrucker/claude · 51 tokens

codebase-analyzer

Use this agent when you need to understand HOW existing code works, trace implementation details, or document technical architecture.

NikiforovAll/claude-code-rules · 27 tokens

explorer-opus

Deeply analyzes complex codebases with comprehensive reasoning to trace subtle execution paths, identify non-obvious root causes, and map intricate architectural dependencies.

Joncik91/ucai · 32 tokens

dead-code-analyzer

Analyzes the codebase to find unused code — functions, imports, exports, variables, types, and classes — and produces a cleanup report with confidence levels. Never auto-deletes; reports findings for user review. Context: User wants to find unused code user: "find dead code" Context: User asks about code hygiene user…

lukaskellerstein/claude-my-marketplace · 160 tokens

team-devops-infra-reviewer

Stage 2 reviewer focused on deployment safety, CI hygiene, IaC, and secrets management.

hazarsozer/crucible-cc · 27 tokens

doctrine-performance-optimizer

Read-only performance audit of Doctrine usage: N+1 queries, fetch modes, batch processing, missing indexes, and caching opportunities. Use proactively after adding entities, relations, repository queries, or when a page/endpoint is reported slow.

dev-toolings/superpowers-symfony · 53 tokens