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 mitodl/agent-kit --skill github-pr-triagegit clone --depth 1 https://github.com/mitodl/agent-kitWrote 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/mitodl/agent-kit/github-pr-triage)<a href="https://agentmods.dev/skills/mitodl/agent-kit/github-pr-triage"><img src="https://agentmods.dev/badge/skills/mitodl/agent-kit/github-pr-triage/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/mitodl/agent-kit/github-pr-triage"><img src="https://agentmods.dev/badge/skills/mitodl/agent-kit/github-pr-triage.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 | $0.00128 | $0.02348 |
| Opus 5.5 | $0.00051 | $0.00939 |
| Sonnet 5.5 | $0.00026 | $0.00470 |
| Haiku 4.5 | $0.00013 | $0.00235 |
Grade A, and why
github-pr-triage 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 22d 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 — 208 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GitHub PR Triage
Turns a sprawling list of open PRs into a report grouped by required action, and
optionally executes that action. The core pattern is
enumerate → enrich → classify → report → (optionally) act, using
gh search prs / gh pr view for data and deterministic jq rules for the
classification — no code archaeology needed, unlike issue triage.
Three deterministic scripts do the mechanical work, so most of this skill is about reading their output correctly and knowing when to escalate to judgment:
| Script | Purpose |
|---|---|
scripts/fetch-prs.sh |
Enumerate open PRs by author across an org |
scripts/enrich-prs.sh |
Fetch review/check/mergeability detail per PR, in parallel |
scripts/classify-prs.sh |
Bucket the enriched PRs by required action |
Action scripts (used only after explicit per-PR confirmation — see Phase 5):
| Script | Purpose |
|---|---|
scripts/request-review.sh |
Request a first-pass review from Copilot and/or Claude; refuses drafts and PRs that already have review activity |
scripts/merge-pr.sh |
Merge an approved, green PR and delete its branch |
Scope: figure out mode before running anything
Two independent choices govern the whole run — resolve both up front, asking the user if either is ambiguous:
- Report or act? A pure status report never mutates anything. Acting means requesting reviews, posting comments, pushing fixes, or merging — all of which are visible to others and some of which are hard to reverse. Default to report-only unless the user clearly asked to take action ("merge the ready ones", "kick off reviews", "clear my backlog").
- Which org/repo, which author? Default author is
@me(the currentghuser) — this skill is built around "my open PRs", matching how the four buckets below are framed (things you, the author, are waiting on or need to act on). If the user wants PRs where they're a reviewer instead, that's a different query (gh search prs --review-requested @me) — say so rather than silently reinterpreting the request.
What ships with it
7 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.
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.
- 22d ago Changed · +7 lines 4c3985f84d15
- 1mo ago First seen · 201 lines · 128 tokens per session scan A 6d034f8b6393
github-pr-triage is a skill published in the GitHub repository mitodl/agent-kit (5 stars, last pushed yesterday), licensed BSD-3-Clause. It adds 128 tokens to every session and 2,348 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
review-implement-phase
Implements triaged review actions, commits focused fixes, and posts Done plus resolves threads. Use when the user wants only the implementation phase of the review-framework workflow.
pr
Draft a PR body from OMC's paper trail — the smallest visual that proves the change, evidence consumed from the verify protocol (never re-collected), reversibility from the ADR test on the diff, glossary language throughout.
merge-batch
Prepare and merge an owner-selected PR batch with contributor credit and exact-candidate admission. Use for a pasted PR list; combined focused validation is diagnostic, and every actual landing still requires its applicable checks and authorization.
otto-pr-review-loop
Run Polly and OCR on a PR until all findings settle.
engram-branch-pr
PR creation workflow for Engram following the issue-first enforcement system. Trigger: When creating a pull request, opening a PR, or preparing changes for review.
verify-behavior
Verify or reproduce visible product behavior by driving the real UI with pi-computer-use's checked tools, requiring verified expect postconditions and durable state evidence for meaningful UI flows. Use when triage needs visual reproduction, implementation needs behavioral proof, review needs interactive confirmation…