pr-push

A pull-request publishing tool that creates or updates a draft pull request (a review request that is not yet published as ready) and retries after fixing pre-commit check failures.

In plain words
What is it for?
Use it to prepare a draft pull request, update an existing one, run repository checks, and fix pre-commit failures until the push succeeds.
Why use it?
It handles the checks that can prevent a branch from being pushed and keeps the pull request in draft form for review before publication.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/wrannaman/agentic-engineering/pr-push
Any agent
npx skills add wrannaman/agentic-engineering --skill pr-push
Clone the repo
git clone --depth 1 https://github.com/wrannaman/agentic-engineering

Made for: Claude Code, Codex.

Per session 22 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,070 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00022 $0.03070
Opus 5 $0.00011 $0.01535
Sonnet 5 $0.00004 $0.00614
Haiku 4.5 $0.00002 $0.00307

Measured 2d ago against content hash ba78d03feb86, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pr-push 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.

skills/git/pr-push/SKILL.md · 329 lines

How it starts

The opening of the file, as written. The whole thing — 329 lines — stays where its author put it; the contents beside it link to each section on GitHub.

PR Push Skill

You are entering the PR push phase. You will create or update a draft PR, automatically fixing any pre-commit hook failures until the push succeeds.

CRITICAL SAFETY RULE: This skill MUST NEVER publish a PR. Always use --draft flag.

Process

Step 1: Pre-flight Checks

Run checks 1-5 in parallel where possible.

  1. Check current branch:

    git branch --show-current
    
    • If on main: Ask user what branch name to use, then create the branch using the detected stack workflow from Step 1.4
    • Otherwise: Continue with current branch
  2. Check if PR already exists (run early — gates later steps):

    gh pr view --json number,state,isDraft,url,title,body,baseRefName 2>/dev/null
    
    • If PR exists: Record PR number, URL, current body, and current base branch. Skip Step 2 (metadata generation) and skip stack submission or git push if remote is already up-to-date. Proceed to update description if needed.
    • If no PR: Will create new draft PR — continue to Step 2.
  3. Check for uncommitted changes:

    git status --porcelain
    
    • If there are uncommitted changes: Warn user and ask if they should be committed first
  4. Detect the active stack workflow using the git-stacks KB partition as the shared source of truth:

    • list_kb_documents for partition git-stacks
    • read_kb_document_by_path for /index.md
    • Run the dedicated-client checks from git-stacks/index.md
    • If a dedicated client is detected, record STACK_CLIENT accordingly
    • If no dedicated client is detected, inspect GitHub base refs and local branch relationships to decide whether this is a native git stack
    • Load the matching KB client doc (/charcoal.md, /git-town.md, or /native-git.md) before pushing or navigating the stack
    • If no stack is confirmed: Use standard git workflow (Step 3b)
  5. Resolve the target PR base from local branch relationships:

    git rev-parse --abbrev-ref --symbolic-full-name @{u} 2>/dev/null
    git log --pretty=format:'%D' | grep -v '^$' | grep -v "^HEAD -> $(git branch --show-current)$" | head -1 | tr ',' '\n' | sed 's/^ *//' | grep -v "HEAD" | grep -v "origin/HEAD" | head -1
    
    • Prefer the detected stack client's local parent metadata when it is available
    • Otherwise, use the git upstream/tracking branch and normalize remote refs like origin/feature/foo to feature/foo
    • Otherwise, fall back to the existing local parent-branch heuristic
    • If multiple sources disagree, or no unambiguous base can be determined, STOP and ask the user
    • Record the resolved branch as TARGET_BASE
    • Never silently fall back to the repository default branch

Read the full file on GitHub · 329 lines

Changes

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.

  1. 2d ago First seen · 329 lines · 22 tokens per session scan A ba78d03feb86

Subscribe to this mod's changes

pr-push is a skill published in the GitHub repository wrannaman/agentic-engineering (2 stars, last pushed 4mo ago), licensed MIT. It adds 22 tokens to every session and 3,070 once invoked, about $0.0001 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens