GitHub Actions Mainline Health Audit

Audit GitHub Actions mainline branch reliability by scoring failure rate, consecutive failures, and stale-success risk for critical workflows.

Security Scan

The skill's code and runtime instructions are consistent with an offline GitHub Actions run-audit tool that reads JSON exports and scores reliability; there are no requests for credentials, no network exfiltration, and no high-risk install steps.

Purpose & Capability

Name/description match behavior: the script and SKILL.md both describe reading GitHub Actions run JSON files, filtering mainline branches, scoring failure/stale-success risk, and emitting text/JSON reports. One minor documentation gap: SKILL.md shows using the `gh` CLI to collect run JSON, but the declared required binaries list only bash and python3 (the script itself does not invoke `gh`).

Instruction Scope

Runtime instructions and the included script operate on local JSON files (RUN_GLOB) and environment-configurable filters; they do not instruct the agent to read unrelated system files, access secrets, or transmit data to external endpoints. The example `gh run view ... > artifacts/...` requires the user to run the GitHub CLI to produce input artifacts, but the skill itself only reads those local files.

Install Mechanism

Instruction-only skill with an included script; there is no install spec, no downloads, and nothing is written to disk by an installer. The script is executed directly by bash/python3, which is proportionate for this purpose.

Credentials

No required environment variables, credentials, or config paths are declared. The many optional environment variables documented are reasonable filters/thresholds for the audit and do not request secrets. Note: using `gh` to collect run JSON (per examples) requires GitHub CLI authentication in the user's environment, but that is external to the skill.

Persistence & Privilege

Skill does not request permanent presence (always=false) and does not modify other skills or system-wide agent settings. It runs on demand and only reads local files provided via RUN_GLOB.

Assessment

This skill appears to be an offline auditor that consumes GitHub Actions run JSON files and emits scores/reports. Things to consider before installing/using: (1) The SKILL.md examples show using the `gh` CLI to produce run JSON files — ensure you have the GitHub CLI and appropriate auth configured if you use that collection method (the skill itself does not request GitHub credentials). (2) Confirm the RUN_GLOB path only matches intended JSON files (avoid accidentally exposing sensitive data). (3) Review the script if you plan to run it in CI to ensure its exit behavior (FAIL_ON_CRITICAL) and output format meet your pipeline expectations. (4) There are no network calls or credential exfiltration in the included code, but always review any third-party skill’s files before running in sensitive environments.

Use this skill to detect unstable workflows on protected branches (main/master/release) before they silently degrade delivery confidence.

What this skill does

• Reads GitHub Actions run JSON exports
• Filters to mainline/protected branches (configurable regex)
• Groups by repository + workflow + branch + event
• Scores risk using:
  • failure rate
  • current consecutive failure streak
  • days since last successful run
• Flags warning/critical groups based on configurable thresholds
• Emits text or JSON output for CI checks and ops dashboards

Inputs

Optional:

• RUN_GLOB (default: artifacts/github-actions/*.json)
• TOP_N (default: 20)
• OUTPUT_FORMAT (text or json, default: text)
• MIN_RUNS (default: 2)
• MAINLINE_BRANCH_MATCH (default: ^(main|master|release.*)$)
• WORKFLOW_MATCH (regex, optional)
• WORKFLOW_EXCLUDE (regex, optional)
• EVENT_MATCH (regex, optional)
• EVENT_EXCLUDE (regex, optional)
• REPO_MATCH (regex, optional)
• REPO_EXCLUDE (regex, optional)
• HEAD_SHA_MATCH (regex, optional)
• HEAD_SHA_EXCLUDE (regex, optional)
• CONCLUSION_MATCH (regex, optional)
• CONCLUSION_EXCLUDE (regex, optional)
• RUN_ID_MATCH (regex, optional)
• RUN_ID_EXCLUDE (regex, optional)
• RUN_URL_MATCH (regex, optional)
• RUN_URL_EXCLUDE (regex, optional)
• FAIL_WARN_PERCENT (default: 20)
• FAIL_CRITICAL_PERCENT (default: 40)
• STALE_SUCCESS_DAYS (default: 7)
• WARN_SCORE (default: 30)
• CRITICAL_SCORE (default: 55)
• FAIL_ON_CRITICAL (0 or 1, default: 0)

Collect run JSON

gh run view <run-id> --json databaseId,workflowName,event,conclusion,headBranch,headSha,createdAt,updatedAt,startedAt,url,repository \
  > artifacts/github-actions/run-<run-id>.json

Run

Text report:

RUN_GLOB='artifacts/github-actions/*.json' \
MAINLINE_BRANCH_MATCH='^(main|release/.*)$' \
HEAD_SHA_MATCH='^[a-f0-9]{7,40}$' \
CONCLUSION_EXCLUDE='^(success)$' \
RUN_ID_MATCH='^50(0[1-5])$' \
MIN_RUNS=3 \
bash skills/github-actions-mainline-health-audit/scripts/mainline-health-audit.sh

JSON output with fail gate:

RUN_GLOB='artifacts/github-actions/*.json' \
OUTPUT_FORMAT=json \
FAIL_ON_CRITICAL=1 \
bash skills/github-actions-mainline-health-audit/scripts/mainline-health-audit.sh

Run with bundled fixtures:

RUN_GLOB='skills/github-actions-mainline-health-audit/fixtures/*.json' \
bash skills/github-actions-mainline-health-audit/scripts/mainline-health-audit.sh

Output contract

• Exit 0 in report mode (default)
• Exit 1 when FAIL_ON_CRITICAL=1 and one or more groups are critical
• Text mode prints summary + ranked mainline-risk groups
• JSON mode prints summary + scored groups + critical group details