Run any Skill in Manus with one click

triage

Stars2

Forks4

UpdatedApril 20, 2026 at 14:26

Analyze the latest failed CI run for a component. Identifies failing jobs and steps, reads logs, determines root cause, and classifies the failure.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Run Skill in Manus

Source

cloudbees-oss

cloudbees-oss/devops-agent-kit

View GitHub Repository View Creator Repositories

Download

Run Skill in Manus

Related occupationsSOC

Based on SOC occupation classification

Network and Computer Systems AdministratorsComputer and Mathematical Occupations·SOC 15-1244

SKILL.md

readonly

name	triage
description	Analyze the latest failed CI run for a component. Identifies failing jobs and steps, reads logs, determines root cause, and classifies the failure.
argument-hint	[component-name]

Triage Failed Run: $ARGUMENTS

Search for the component using components_search
Pull recent runs using runs_list(limit=10)
Filter for FAILED runs. Skip any with workflow names containing "code_scan"
For the latest failed run: a. automation_jobs_list — list all jobs and their statuses b. For each failed job, identify the failing step c. logs_list — attempt to pull logs from the failing step NOTE: If this returns a 500 error, proceed without logs. Analyze based on step names, statuses, and durations instead. d. workflow_get_content — read the workflow YAML to understand what each step does and identify configuration issues
Classify the failure:
- CONFIG: Workflow/pipeline configuration error (missing inputs, invalid action references, YAML syntax). Look for validation errors.
- CODE: Compilation failure or test failure caused by a code change. Check if earlier steps (build, compile) passed.
- SECURITY: Security scan gate failure (Gitleaks, SAST, SCA). Cross-reference with security_findings_summary_get.
- ENVIRONMENT: Infrastructure issue — timeouts, connection refused, resource exhaustion, runner unavailability.
- KNOWN: Matches a pattern from previous triage (same step, same error).
Enrich with security context: a. security_findings_summary_get — check if the component has open security findings that may be related to the failure
Present a structured triage report:
- Run ID, workflow name, trigger, timestamp
- Step-by-step results table (step name, status, duration)
- Root cause analysis with specific evidence
- Classification (CONFIG/CODE/SECURITY/ENVIRONMENT/KNOWN)
- Fix recommendation
Offer next actions:
- File a Jira ticket for the issue (/file-ticket)
- Post the summary to Slack
- Comment on the relevant GitHub PR
- If the fix is a workflow config change, offer to apply it via workflow_update_content (with user confirmation)

More from this repository

same repository

ci-health

cloudbees-oss/devops-agent-kit

Generate a CI health report for a component or the entire organization. Uses controller analytics, run history, and security posture.

2026-04-202

file-ticket

cloudbees-oss/devops-agent-kit

Create a Jira ticket from a build failure, security finding, or improvement recommendation.

2026-04-202

flag-rollout

cloudbees-oss/devops-agent-kit

Manage feature flags — list, inspect, enable, disable, or create flags. Supports progressive rollout workflows.

2026-04-202

pipeline-overview

cloudbees-oss/devops-agent-kit

Show a unified view of all CI/CD pipelines across Jenkins, GitHub Actions, and CloudBees Workflows. The cross-tool picture no single tool can provide.

2026-04-202

release-status

cloudbees-oss/devops-agent-kit

Check release readiness across all components. Verifies CI status, security gates, and feature flag states for a go/no-go decision.

2026-04-202

security-scan

cloudbees-oss/devops-agent-kit

Analyze security posture across components. Shows vulnerabilities by severity, scanner type, SLA status, and remediation priority.

2026-04-202

name	triage
description	Analyze the latest failed CI run for a component. Identifies failing jobs and steps, reads logs, determines root cause, and classifies the failure.
argument-hint	[component-name]

Triage Failed Run: $ARGUMENTS

Search for the component using components_search
Pull recent runs using runs_list(limit=10)
Filter for FAILED runs. Skip any with workflow names containing "code_scan"
For the latest failed run: a. automation_jobs_list — list all jobs and their statuses b. For each failed job, identify the failing step c. logs_list — attempt to pull logs from the failing step NOTE: If this returns a 500 error, proceed without logs. Analyze based on step names, statuses, and durations instead. d. workflow_get_content — read the workflow YAML to understand what each step does and identify configuration issues
Classify the failure:
- CONFIG: Workflow/pipeline configuration error (missing inputs, invalid action references, YAML syntax). Look for validation errors.
- CODE: Compilation failure or test failure caused by a code change. Check if earlier steps (build, compile) passed.
- SECURITY: Security scan gate failure (Gitleaks, SAST, SCA). Cross-reference with security_findings_summary_get.
- ENVIRONMENT: Infrastructure issue — timeouts, connection refused, resource exhaustion, runner unavailability.
- KNOWN: Matches a pattern from previous triage (same step, same error).
Enrich with security context: a. security_findings_summary_get — check if the component has open security findings that may be related to the failure
Present a structured triage report:
- Run ID, workflow name, trigger, timestamp
- Step-by-step results table (step name, status, duration)
- Root cause analysis with specific evidence
- Classification (CONFIG/CODE/SECURITY/ENVIRONMENT/KNOWN)
- Fix recommendation
Offer next actions:
- File a Jira ticket for the issue (/file-ticket)
- Post the summary to Slack
- Comment on the relevant GitHub PR
- If the fix is a workflow config change, offer to apply it via workflow_update_content (with user confirmation)