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 ErikaAX08/erikas-skills --skill create-pull-requestgit clone --depth 1 https://github.com/ErikaAX08/erikas-skillsWrote 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/erikaax08/erikas-skills/create-pull-request)<a href="https://agentmods.dev/skills/erikaax08/erikas-skills/create-pull-request"><img src="https://agentmods.dev/badge/skills/erikaax08/erikas-skills/create-pull-request.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 5.1 | $0.00138 | $0.04031 |
| Opus 5 | $0.00069 | $0.02015 |
| Sonnet 5 | $0.00028 | $0.00806 |
| Haiku 4.5 | $0.00014 | $0.00403 |
Grade A, and why
create-pull-request 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 8d 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 — 360 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CREATE-PULL-REQUEST — Pull Request Creation Skill
Purpose: operational rulebook for drafting the title and description of a Pull Request (or Merge Request) grounded in the repository's actual git diff — always following first the template the repository itself defines, and only in its absence the improved default template in this file.
Golden rule: the repository's template wins. This file is the default, never an override. If the repo defines its own PR template, you fill that one and do not mix it with the one here.
0. How to Use This File
- Before writing anything, determine the real scope of the changes in git (§3, Step 1): the description must be anchored to the diff, not to memory.
- Look for the repo's template (§3, Step 2). If it exists: use it exactly as it is. If not: use the default template in §4 (copy its shape, don't improvise).
- Fill in only what you can verify — from the diff, from commands you ran, or from what the user said explicitly. Anything unverifiable is omitted or marked as pending. It is never invented.
- Deliver per §3, Step 6: the result is always saved as a numbered Markdown
file in the
.pr-suggestions/folder, and the PR is never opened on your own.
1. Non-Negotiables
- Never fabricate issue numbers, file paths, changes, test results, executed
commands, screenshots, or breaking-change claims without evidence. A
Closes #123that doesn't come from the repo or the user is a serious error. - The repo's template always wins. If the repository defines its own PR template,
or its
CONTRIBUTING.mdspecifies the format, follow that. This file is the default for repos with no stated convention, not an override. - Every "Change" listed must exist in the actual diff. Describing a change that was never made is fabrication — each bullet under Changes made must map to a real behavioral/functional difference in the diff.
- Never report tests you did not run. Write only results whose output you read in
the session. If not run, write
Not run — pending in CIand leave the checkbox unchecked. - No AI co-authorship. No
Co-Authored-By: <AI>or any mention of this kind in the PR body. - No placeholders in the final output. An empty section is removed (no dangling
[complete]or<!-- ... -->); an empty bullet is deleted. - Use the repository's language. Write the description in the repo's dominant language (documentation, commit history, previous PR titles). One language per PR; never mix.
- Never reinterpret already established facts. If a commit says
fix(auth): ..., the PR reflects exactly that; don't rewrite it as something broader or different to "sound better." - Always write the PR description to a Markdown file in
.pr-suggestions/. The description is never delivered only in the terminal or saved anywhere else; the folder (and file) is created if missing. - Number every PR file sequentially. Files are
PR-001.md,PR-002.md, … — one per suggestion, never overwritten and never reused. The highest number is always the most recent PR created. - Never constrain paragraph width. No manual line wrapping/breaks inside paragraphs (e.g. hard-wrapping at 80 columns); write each paragraph as a single continuous line and let the renderer wrap it.
- Always suggest a branch name. Every PR description includes a proposed branch
name (§3, Step 3), following the repo's own naming convention when one is
detectable, or the
<type>/<scope>-<short-description>kebab-case default otherwise. Never skip it, and never invent one unrelated to the actual change.
What ships with it
1 file 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.
- 8d ago First seen · 360 lines · 138 tokens per session scan A 1a1bcc0e25d5
create-pull-request is a skill published in the GitHub repository ErikaAX08/erikas-skills (5 stars, last pushed 8d ago), licensed MIT. It adds 138 tokens to every session and 4,031 once invoked, about $0.0007 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
gentle-ai-collab-perfect
Trigger: contributing to Gentleman-Programming/gentle-ai as an external collaborator. Strict issue-first workflow, honest PR bodies, contributor-vs-maintainer scope, chained-PR strategy, verification protocol, docstring coverage. Load whenever the active repo is Gentleman-Programming/gentle-ai and any part of the…
branch-pr
Create Gentle AI pull requests with issue-first checks. Trigger: creating, opening, or preparing PRs for review.
work-unit-commits
Plan commits as reviewable work units. Trigger: implementation, commit splitting, chained PRs, or keeping tests and docs with code.
chained-pr
Trigger: PRs over 400 lines, stacked PRs, review slices. Split oversized changes into chained PRs that protect review focus.
revert
Git-aware revert that understands Draft tracks, phases, and tasks. Safely undo work at task, phase, or track level. Use when the user asks to 'revert this track', 'undo a phase', 'revert task X', or says 'roll back the last task', 'undo this work'.
upload
Pre-upload gate for track handoff. Verifies review status, HLD approvals, deploy checklist, and validator chain before git upload or PR submission.