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.
npx skills add openshift-eng/ai-helpers --skill investigate-ci-reliabilitygit clone --depth 1 https://github.com/openshift-eng/ai-helpersWrote 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.
[](https://agentmods.dev/skills/openshift-eng/ai-helpers/investigate-ci-reliability)<a href="https://agentmods.dev/skills/openshift-eng/ai-helpers/investigate-ci-reliability"><img src="https://agentmods.dev/badge/skills/openshift-eng/ai-helpers/investigate-ci-reliability/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.
<a href="https://agentmods.dev/skills/openshift-eng/ai-helpers/investigate-ci-reliability"><img src="https://agentmods.dev/badge/skills/openshift-eng/ai-helpers/investigate-ci-reliability.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00048 | $0.01561 |
| Opus 5.5 | $0.00019 | $0.00624 |
| Sonnet 5.5 | $0.00010 | $0.00312 |
| Haiku 4.5 | $0.00005 | $0.00156 |
Grade A, and why
investigate-ci-reliability 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 9d 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.
How it starts
The opening of the file, as written. The whole thing — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CI reliability investigation
Investigate actual failed job runs, demonstrate the underlying defects, and deliver
an issues/ directory containing only independently validated, currently applicable
fixes. This package contains its own collectors, review contract, and exporter.
Use the prow-job-analysis skill from the ci plugin for failed-job debugging;
no prior investigation repository is required.
Inputs and defaults
release: required OpenShift release, inferred from the request when stated.- Window: last 24 hours, frozen at collection start; explicit UTC start/end supported.
- Scope: all jobs in the release plus presubmits. Presubmits are the separate
Sippy
Presubmitspopulation, not implicitly restricted to merged PRs or that release. - Optional scope: release-only, presubmit-only, or verified payload-blocking runs. Optional exact job, job substring, and variant filters narrow either population.
max-issues: 10 distinct validated mechanisms, a ceiling rather than a quota.max-candidates: 100;time-budget-minutes: 120;max-agents: 4. These bound exploration, including review. Honor smaller user limits.- Workspace/output: user-selected paths, otherwise
~/tmp/ci-reliability/<UTC timestamp>and itsvalidated/subdirectory. Scratch defaults to~/tmp/ci-reliability. Keep raw downloads bounded and separate from curated handoff evidence.
Invoke /investigate-ci-reliability 5.1 --max-issues 10 directly. The skill accepts
these inputs in natural language or flags. Resolve bundled script paths relative to this SKILL.md, regardless of cwd.
The bundled scripts need Python 3.10+ and public HTTPS access. JUnit parsing requires
Expat 2.7.2+ in that Python runtime. The investigation also requires the
prow-job-analysis skill from the ci plugin.
Start and collect
Read collection controls for exact collector flags, verified blocking selection, pagination behavior, and completeness limits.
What ships with it
10 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/artifact-helpers.md 4.8 KB
- references/collection.md 9.7 KB
- references/evidence-contract.md 4.8 KB
- references/proof-review.md 5.4 KB
- scripts/collect_runs.py 23 KB runs code
- scripts/prow_artifacts.py 22 KB runs code
- scripts/reliability.py 14 KB runs code
- scripts/test_collect_runs.py 14 KB runs code
- scripts/test_prow_artifacts.py 16 KB runs code
- scripts/test_reliability.py 7.4 KB runs code
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.
- 9d ago Changed · +5 lines d7a6bdbe7604
- 16d ago First seen · 126 lines · 48 tokens per session scan A 7e034067a803
investigate-ci-reliability is a skill published in the GitHub repository openshift-eng/ai-helpers (121 stars, last pushed today), licensed Apache-2.0. It adds 48 tokens to every session and 1,561 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-06.
Other skills, from other repositories
repair-github-ci
Fix failing GitHub CI / Actions checks via fixgithubprci and push to the existing PR head, or fix a branch's failing CI via a linked repair worktree.
mcore-create-issue
Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.
debug-task
Diagnose and fix moon tasks that are broken, misconfigured, or behaving unexpectedly. Use this skill when a moon task is failing, not running, skipped, hanging, producing stale or wrong output, cached when it shouldn't be, re-running every time, or missing outputs after a cache hit. Also covers pipeline hangs, tasks…
operating-github-ci-fixer
Use when the user asks OpenSRE to fix failing GitHub CI, GitHub Actions checks, failing pull request checks, a broken PR branch, or CI on a named branch such as main.
meta-long-running-build-watchdog
Watches a long-running command via tmux, lets sub-agent diagnose failures and propose a fix, and records the diagnosis to memory. Designed for overnight model fine-tunes, CI image builds, or repeated regression suites that may fail intermittently.
gh-fix-ci
Use when a user asks to debug or fix failing GitHub PR checks that run in GitHub Actions; use gh to inspect checks and logs, summarize failure context, draft a fix plan, and implement only after explicit approval. Treat external providers (for example Buildkite) as out of scope and report only the details URL.