Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/thettwe/nyannnpx agentmods add skills/thettwe/nyann/prWrote 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/thettwe/nyann/pr)<a href="https://agentmods.dev/skills/thettwe/nyann/pr"><img src="https://agentmods.dev/badge/skills/thettwe/nyann/pr/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/thettwe/nyann/pr"><img src="https://agentmods.dev/badge/skills/thettwe/nyann/pr.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 | $0.00184 | $0.01605 |
| Opus 5.5 | $0.00074 | $0.00642 |
| Sonnet 5.5 | $0.00037 | $0.00321 |
| Haiku 4.5 | $0.00018 | $0.00161 |
Grade A, and why
pr 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 1mo 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 — 150 lines — stays where its author put it; the contents beside it link to each section on GitHub.
pr
Wraps bin/pr.sh. Two modes:
- Context-only (
--context-only). Collects branch + commit range- suggested title without any network calls. The gh auth check is
skipped entirely in this mode (the script exits with the JSON
summary before reaching the gh guard), so context-only also works
when
ghisn't installed or authenticated. Use this when the user wants you to synthesize a PR title and body from the commits before shipping.
- suggested title without any network calls. The gh auth check is
skipped entirely in this mode (the script exits with the JSON
summary before reaching the gh guard), so context-only also works
when
- Create (default). Pushes the current branch with
-utracking and invokesgh pr createwith your title + body. Returns the PR URL.
0. Drift check (quick, non-blocking)
Run bash bin/session-check.sh --flow=pr. If it produces output,
surface the line to the user verbatim. Do not block the flow.
1. Pre-flight
- Verify
ghis installed and authenticated.bin/pr.shdoes this and emits{skipped, reason}with reasongh-not-installedorgh-not-authenticated. If you see a skip, relay it cleanly; don't try to installghfor the user. - Verify the current branch is not
main/master. If it is, refuse and suggest/nyann:branchto create a feature branch first. - Verify there are commits ahead of the base. If
ahead == 0, tell the user there's nothing to PR and suggest they commit first.
Pre-action guards
bash bin/pre-action-guard.sh --flow pr --target <cwd> [--profile <resolved-profile.json>]
| Exit | Meaning | Action |
|---|---|---|
| 0 | Pass (or advisory warnings) | Surface advisory message lines, continue |
| 3 | Critical guard failed | Refuse to open the PR unless the user passes --skip-guards. Re-ask via AskUserQuestion. |
| 4 | Confirm-severity guard failed | AskUserQuestion: "Proceed despite the warnings?" — explicit confirm required. |
Built-in PR guards include branch-pushed (advisory if upstream missing
or ahead/behind) and wip-commits (advisory if the commit range still
contains WIP/fixup commits). Profiles can promote either to confirm or
critical via guards.pr in the profile.
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.
- 1mo ago First seen · 150 lines · 184 tokens per session scan A 24d50f84273f
pr is a skill published in the GitHub repository thettwe/nyann (7 stars, last pushed 6d ago), licensed MIT. It adds 184 tokens to every session and 1,605 once invoked, about $0.0007 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 2026-09-03.
Other skills, from other repositories
pr-create
Creates a pull request from current changes, monitors GitHub CI, and debugs any failures until CI passes. Activate when the user says "create pr", "make a pr", "open pull request", "submit pr", "pr for these changes", or wants to get their current work into a reviewable PR. Assumes the project uses git, is hosted on…
release
CONTRIBUTOR TOOL - Cut a plugin release: bump plugin.json version, finalize CHANGELOG, update README if needed, gate on make ci, commit, tag vX.Y.Z, and create the GitHub release. Use when shipping a new plugin version. NOT distributed.
catchup
Summarize and review what changed while you were away. Use after a weekend, vacation, or flight to check missed PRs, git commits, Linear tickets, and meetings — one prioritized brief, not a firehose.
watch-pr
Watch an Elixir/Phoenix PR for new review comments (bot + human) and CI results via a background watcher that wakes Claude only on real events. Use after opening a PR or pushing, while waiting on CI or reviewers.
pr-review
Address feedback left on a GitHub pull request: fetch unresolved review threads, make agreed Elixir/Phoenix code fixes, reply, and resolve. Use for a PR URL/number or reviewer comments. NOT for pre-PR review, findings triage, or CI monitoring.
create-release-checklist
Create a release checklist and GitHub issue for an R package. Use when the user asks to "create a release checklist" or "start a release" for an R package.