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 antropos17/Aegis --skill ci-monitorgit clone --depth 1 https://github.com/antropos17/AegisWrote 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/antropos17/aegis/ci-monitor)<a href="https://agentmods.dev/skills/antropos17/aegis/ci-monitor"><img src="https://agentmods.dev/badge/skills/antropos17/aegis/ci-monitor/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/antropos17/aegis/ci-monitor"><img src="https://agentmods.dev/badge/skills/antropos17/aegis/ci-monitor.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 | $0.00029 | $0.00424 |
| Opus 5.5 | $0.00012 | $0.00170 |
| Sonnet 5.5 | $0.00006 | $0.00085 |
| Haiku 4.5 | $0.00003 | $0.00042 |
Grade A, and why
ci-monitor 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 today.
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.
What it actually says
Monitor AEGIS CI
Resolve the repository from the checkout's remote or the supplied PR URL. Identify the target PR and head commit before reading results. For a PR, use its check rollup; a workflow may run on a synthetic merge commit, so filtering runs only by the branch head can miss it.
Useful read-only commands, with placeholders replaced by observed values:
gh pr view <number> --repo <owner/repo> --json headRefOid,statusCheckRollup,url
gh pr checks <number> --repo <owner/repo> --json name,state,bucket,link
gh run list --repo <owner/repo> --commit <sha> --json databaseId,headSha,status,conclusion,url
gh run view <run-id> --repo <owner/repo> --json status,conclusion,headSha,jobs,url
A run is finished only when status is completed. Report success only from a successful conclusion. Queued, waiting and pending are unfinished; cancellation, timeout, skipped jobs and missing checks are not proof of success. gh pr checks exit code 8 means pending. An API/authentication failure is unavailable evidence, not a failed build.
Match required checks against AGENTS.md, the current CI workflow and available branch requirements. If the PR head changes, refresh the target and do not reuse results from the old revision. On failure, inspect the failed job and relevant logs.
For active watching, use bounded waits and stop at the requested outcome or a result requiring action. A routine post-push check does not create a recurring automation. For explicitly requested recurring monitoring, use the host's available scheduler; /loop is host-specific, not a universal command. Save the target, stopping condition and notification intent; stay quiet while nothing meaningful changes.
Fetch stars, forks or issue counts only when requested. Report measured changes without invented growth targets or severity thresholds.
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.
- today First seen · 24 lines · 29 tokens per session scan A 3d0675ca70ff
ci-monitor is a skill published in the GitHub repository antropos17/Aegis (157 stars, last pushed today), licensed MIT. It adds 29 tokens to every session and 424 once invoked, about $0.0001 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 2026-10-17.
Other skills, from other repositories
fix-ci
Fix the failed CI checks for the current session. Use when the user requests a CI fix via the Fix Checks button in the Changes toolbar.
analyzing-github-ci-performance
Computes a CI/CD metrics table from raw GitHub Actions records for one repository over the last 30 days, including failure rates and developer waiting time. Use for historical CI performance questions or the first-experience repository demo. For currently failing checks, use reporting-github-ci-failures.
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.
scheduling-github-ci-repairs
Sets up ongoing local monitoring of one repository's open pull requests, automatically editing, testing, and pushing fixes for failing GitHub Actions checks. Offers a disposable private-repository demonstration. Use for recurring PR repair or the local CI onboarding demo.
delegating-github-ci-repairs
In the interactive shell, prepare the hosted gateway and delegate a bounded GitHub PR repair or remote demo. Use for remote CI repair onboarding.
analyzing-local-repositories
Shows what the user's local repositories say about how they work, read from git and CI files on this machine with no GitHub token or network: follow-up fixes, CI trial and error, AI co-authorship, workflow hygiene, tests and loose ends. Use when GitHub is not connected or cannot be read during onboarding, when the…