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 heristop/leclap --skill pr-creationgit clone --depth 1 https://github.com/heristop/leclapWrote 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/heristop/leclap/pr-creation)<a href="https://agentmods.dev/skills/heristop/leclap/pr-creation"><img src="https://agentmods.dev/badge/skills/heristop/leclap/pr-creation/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/heristop/leclap/pr-creation"><img src="https://agentmods.dev/badge/skills/heristop/leclap/pr-creation.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 | $0.00050 | $0.00981 |
| Opus 5.5 | $0.00020 | $0.00392 |
| Sonnet 5.5 | $0.00010 | $0.00196 |
| Haiku 4.5 | $0.00005 | $0.00098 |
Grade A, and why
pr-creation 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 2d 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.
Pull requests in this repo
The rule
A PR merges as one squash commit whose message is the PR title alone. Everything else (the description, the evidence, the review) lives on GitHub, never in git log.
1. Commits on the branch
- Format: Conventional Commits, lowercase, subject only, no body:
feat(engine): add the burst fx primitive. Scopes in use:engine,core,web,expo,cli,mcp,examples,skills,samples,pr-media,changelog,release. - Attribution: none, in commits or in PR bodies. No
Co-Authored-By, no "Generated with" line. - Consent: commit, push, open or edit a PR only once the user has asked for it or approved it. Approval for one PR doesn't carry over to the next.
- Small changes: a changelog-only or docs-only catch-up may go straight to
maininstead of a PR. Ask which.
2. The PR title
- It becomes the squash commit's subject, so write it as one: conventional, lowercase, under ~72 characters, the user-visible outcome (
feat(engine): keyframe tracks on image, animation and html layers). - One concern per PR. If the title needs "and" twice, split the PR.
3. The description
Readers are reviewers on GitHub, not git log. Keep it scannable:
- The reason first, in plain words: the defect, or what users couldn't do. Never lead with the change list, stars or download counts.
- The evidence right after, when the diff touches
apps/leclap-weborapps/leclap-expo: before/after snapshots and a LeClap-made video, playing inline. Follow the pr-evidence skill (attach-video.jsthrough the user's browser). With no visible change, say so in one line:No visible change: <why>. - The change: API, how it lowers, on-device status, one short section each.
- Tests, with what was measured, not just "tests pass".
- Limits and follow-ups.
Pin raw media URLs to the full commit SHA. Name videos <project>-<ticket>-evidence.mp4.
4. CI
- Watch it until it finishes:
gh pr checks <n>, then read the failing log withgh run view <run-id> --log-failed. - Fix the cause, not the guard. A size budget, golden or lint rule that trips on every PR is a design signal: fix the structure, and raise a budget only with a comment saying what grew and why.
- A test that only fails locally on a loaded host (timeouts) isn't a CI failure; say so and move on.
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.
- 2d ago First seen · 66 lines · 50 tokens per session scan A 5cbb94feea8e
pr-creation is a skill published in the GitHub repository heristop/leclap (34 stars, last pushed yesterday), licensed MIT. It adds 50 tokens to every session and 981 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 2026-10-14.
Other skills, from other repositories
github
GitHub patterns using gh CLI for pull requests, stacked PRs, code review, branching strategies, and repository automation. Use when working with GitHub PRs, merging strategies, or repository management tasks.
create-pr
Creates a GitHub PR with a Linear-ticket-prefixed title and a decision-led, narrative description for Prisma 8. Use when the user wants to create a pull request, open a PR, or submit changes for review.
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.
review-triage-phase
Produces canonical review actions from fetched review state and renders action markdown. Use when the user wants only triage/action-planning for the review-framework workflow.
multiline-commit-messages
Use single-quoted strings for multiline git commit messages in the Shell tool. Prevents heredoc escaping failures that produce garbled commit messages.
qiaomu-cut
A workflow for turning a brief video idea into a planned, sourced, edited, rendered, and checked video project.