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 agentmods add skills/kouroshez/coding-os/pr-mode-drivernpx skills add kouroshez/coding-os --skill pr-mode-drivergit clone --depth 1 https://github.com/kouroshez/coding-osWrote 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/kouroshez/coding-os/pr-mode-driver)<a href="https://agentmods.dev/skills/kouroshez/coding-os/pr-mode-driver"><img src="https://agentmods.dev/badge/skills/kouroshez/coding-os/pr-mode-driver.svg" alt="Measured on agentmods" 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 | $0.00080 | $0.01162 |
| Opus 5 | $0.00040 | $0.00581 |
| Sonnet 5 | $0.00016 | $0.00232 |
| Haiku 4.5 | $0.00008 | $0.00116 |
Grade A, and why
pr-mode-driver scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
pr-mode Driver Skill
You are driving the pr-mode autonomous loop for a consumer repo (COS_GIT_WORKFLOW=pr). One agent owns one task → one agents/* branch → one worktree → one PR. After cos pr submit, your job is to poll CI and act — never merge by hand, never spin in-process.
Skip entirely in trunk mode (coding-os itself, and any default repo — pr-mode is consumer-only, default OFF). Full spec: pr-workflow.md §§ 4–10.
The one signal
cos pr status --branch "$BRANCH" --json # → {"ci_rollup": "merged|red|queued|pending|passing|passing-unarmed|closed|none"}
ci_rollup collapses gh's statusCheckRollup + PR state + whether auto-merge is actually armed into one value. Branch on it — nothing else; the STOP decision needs no memory of the prior cos pr submit.
The loop — ONE pass per turn (never busy-wait)
ci_rollup |
Action |
|---|---|
merged |
cos pr cleanup --task "$TASK" — removes the worktree + branch. Done. |
red |
Heal (below). |
passing |
Green AND auto-merge is armed — it will land itself. Re-poll next turn; do nothing else. |
passing-unarmed |
Green but nothing will auto-land it (autonomy=draft, no required check, or a draft PR). STOP — needs a human merge (below). |
pending |
Checks still running. Re-poll next turn. |
queued |
In the GitHub merge queue — the queue will land it (or eject it) on its own. Re-poll next turn; do not re-submit or re-arm. An ejected PR surfaces as red (only that PR is healed; followers keep merging). |
none |
No PR yet → cos pr submit (or cos pr open if there's no worktree). |
closed |
Closed unmerged (human/abandoned). Stop and surface to the user; do not auto-reopen. |
passing-unarmed → STOP, don't spin. The signal itself says nothing will land this PR on its own — you do NOT need to recall the prior cos pr submit, so this holds across /clear, /compact, and a reaper-recovered fresh session. Do NOT re-poll — at autonomy=draft that would loop forever. Stop and tell the user: "PR #N is green and needs a human merge (auto-merge isn't armed — autonomy=draft, or no required status check on the integration branch) — merge it, or set autonomy_level=auto_merge with a required check in Hub Config→Git."
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.
- 3d ago First seen · 66 lines · 80 tokens per session scan A 925a93c2e919
pr-mode-driver is a skill published in the GitHub repository kouroshez/coding-os (6 stars, last pushed 3d ago), licensed Apache-2.0. It adds 80 tokens to every session and 1,162 once invoked, about $0.0004 per session on Opus 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
gate-probe
Prove a gate before trusting it. Inventory violations, plant a defect, verify RED, restore GREEN, prevent idle scans, and confirm automatic wiring.
mykg-github-pages
Set up and maintain the GitHub Pages site for the mykg repo (SenolIsci/mykg) — a purpose-built pages/ folder (landing page adapted from README.md, blog posts, diagrams), built by a GitHub Actions workflow that runs Jekyll and deploys the result to a gh-pages branch. Use whenever the user wants to publish project…
ci-validation-gates
Defensive CI/CD patterns: semver validation, token checks, retry logic, draft detection — earned from v0.8.22.
graph-mutation-plan
Cookbook for composing an applygraphmutations plan — stable entitykey patterns, the canonical label/edge vocabulary, evidence/invalidation/confidence discipline, and a worked example. Load this when building a non-trivial mutation plan.
chinese-git-workflow
国内 Git 平台配置参考——Gitee、Coding.net、极狐 GitLab、CNB 的 SSH/HTTPS/凭据/CI 接入差异与镜像同步配置。仅在用户显式 /chinese-git-workflow 时调用，不要根据上下文自动触发。.
azsdk-common-pipeline-analysis
Analyze Azure SDK CI/CD pipeline failures into a structured diagnosis, and define the required output format. Load this skill before calling azsdkanalyzepipeline, which returns raw failure data that this skill interprets and formats. USE FOR: "pipeline failed", "build failure", "CI check failing", "tests failing in…