Q Branch

    1

    Installs four MIT-licensed gadgets that make coding agents cheaper and more reliable, using hooks instead of longer prompts.

    Free

    1 useSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Q Branch

    Example session with this skill installed

    Set up groundhog in this repo and show me what my agents keep getting wrong.

    • Read your context and instructions
    • Compiled the q branch

    Summary

    Installed groundhog, mined this repo's session history, and turned the most-repeated mistake into an enforced lesson. Output below is from a real run on a repo with ~1,200 sessions of history.

    1. Install

    claude plugin marketplace add grexican/qbranch
    claude plugin install groundhog@qbranch
    

    2. Mine the history

    $ groundhog scan
    1187 sessions, 81291 shell calls: 21 lesson candidates, 50 sequences, 32 friction shapes, 100 repeats.
    Next: groundhog report
    

    3. Read what it found

    $ groundhog report
    Lessons: the same error, then the same workaround (accept with: groundhog accept L<n>)
      L1  x14 in 14 sessions  pnpm -s db:seed --env  =>  pnpm db:seed --env dev
    
    Sequences run by hand again and again (draft a runbook with: groundhog draft S<n>)
      S1  71 sessions  dotnet build P -v  >  dotnet test P --no-build
    
    Friction: repeated commands that fail a lot
      F1  97% of 39 fail  npm run verify
    

    That first line is the same lesson re-learned in

    14 separate sessions: the agent runs the command one way, it fails, and it then works the other way. The sequence in 71 sessions is a verify script nobody wrote.

    4. Accept the lesson

    $ groundhog accept L1
    L1 added (deny-once): when /pnpm\s+-s\s+db:seed\s+--env/
       -> this form fails here; what worked: pnpm db:seed --env dev
    

    It lands in groundhog.lessons.json, which you commit, so the whole team's agents get it.

    What happens next session

    The lesson is injected at session start. If the agent reaches for the old command anyway, a PreToolUse hook blocks it

    once, with the fix, then lets it through on a retry, so a false positive can never deadlock the run.

    Next steps

    • groundhog draft S1 turns that 71-session sequence into a runbook skill.
    • Review before accepting: mined lessons go stale. On one machine the top mined lesson was "use python3", but by then both python and python3 were broken and only py -3 worked.
    • groundhog stats later tells you whether an accepted lesson actually stopped recurring.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Q Branch is four open-source gadgets for coding agents. This skill installs the ones you choose and explains how to drive them. Everything is MIT-licensed and public: https://github.com/grexican/qbranch

    They share one idea: guarantee steps with hooks instead of longer prompts. A rule in CLAUDE.md or AGENTS.md costs tokens in every session and still gets skipped. A hook costs nothing until the moment it applies, and then it fires every time.

    The gadgets

    • headjack — Skills arrive when a task needs them. The library is indexed and searched per prompt instead of preloaded. This matters because large libraries lose their skill descriptions from the listing (measured on Claude Code 2.1.285: roughly 8 skills keep a description on Haiku, 105 on Sonnet), so a plain-language request can no longer be matched to a skill the agent technically has.
    • middleout — One plan over your MCP servers: chained calls, one-to-many fan-out, automatic pagination, and only the fields you select come back.
    • moneyball — Measures every subagent run from the transcripts (model, cost, and whether it actually worked), then graduates a kind of task to a cheaper model once the evidence supports it. It judges cost per completed task, so a cheap model that needs retries is not counted as cheap.
    • groundhog — Mines transcripts for the lessons agents keep re-learning, keeps accepted ones in a committed ledger, and enforces them with a hook that blocks the known-bad command once, with the fix.

    Honest limits

    • groundhog needs session history to mine. A fresh machine has nothing to learn from yet.
    • moneyball needs at least 10 runs of a task class before recommending anything, and its automatic model routing applies to Claude models; for other vendors it reports advice.
    • headjack helps most with a large library. With a handful of skills, your agent's built-in preloading is fine, and the honest answer is to keep using it.
    • middleout earns its place on per-item fan-out over chatty APIs. For a one-off call, a raw tool call is simpler. A plan runs under a single approval, so a fan-out of 100 items makes 100 calls; it is exactly as safe as the tools the plan names.

    Requirements

    Node 22.13+, or the native binary (Windows, macOS, Linux). Works with Claude Code, Cursor, Codex, Gemini CLI, Copilot and Windsurf. Where an agent cannot do something, that feature turns off rather than half-working, and each gadget's README lists exactly what works where.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 29 days ago

    • Free with an Agensi account

    Requires

    GitHub
    Node.js 22.13+ or native binary
    Q Branch
    Git
    macOS

    Permissions

    Terminal / Shell
    Network Access

    Allowed Hosts

    raw.githubusercontent.com
    Listed29 days ago

    What's inside

    Frequently Asked Questions

    Installs four MIT-licensed gadgets that make coding agents cheaper and more reliable, using hooks instead of longer prompts.

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use Q Branch on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    New in LLM Ops