More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    The Quartermaster: quests and decisions your agent keeps

    1

    Every task gets a goal, done checks, decisions with their reason, and a wrap-up you grade. Nothing gets forgotten.

    $12

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    The Quartermaster: quests and decisions your agent keeps

    The Quartermaster: quests and decisions your agent keeps

    Example session with this skill installed

    Add Spanish VAT to the cart: 21% of the total, rounded to cents. VAT must round half up, never banker's rounding.

    • Read your context and instructions
    • Compiled the the quartermaster

    Quest vat-on-cart: 5 checks agreed, 3 decisions saved with reasons, 'round half up' added to the rule book, sub-agent review recorded, wrap-up GREEN on HEAD waiting for your grade.

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

    About this skill

    Keeps the ledger: every quest, every decision, every why.

    Long agent sessions lose track. A week later nobody remembers why a choice was made, which checks proved the work, or which of five tasks is waiting on you.

    The Quartermaster gives your coding agent a simple workflow, kept in plain files in your repo (.guild/):

    1. Every task is a quest with a goal and "done" checks you approve before any code.
    2. Every decision is saved when it is made, with its reason. Rules that apply beyond one task go into a dated rule book, and the agent reads it before the next quest.
    3. Questions only you can answer show up on the board until you answer them or park them with a date. A quest cannot close while one is pending.
    4. A board with four fixed sections: Needs you, Underway, Next, Landed. One look, no digging.
    5. A weekly look back: rules --since 7d lists what was decided and why.
    6. Scouts investigate in a scratch worktree and report findings. The real fix starts clean, so debug code never leaks into a PR.
    7. A wrap-up with check results on HEAD, what changed, decisions, the review and open questions. You grade it and pick the next step.

    One Python script, standard library only. No server, no account, nothing leaves your machine. Works with Claude Code, Codex, Cursor and any agent that can run a shell.

    Built from guild, an open-source (MIT) multi-agent workflow, cut down to what works inside a single skill.


    Part of The Guild, agent skills by Sesisnando Neto: The Oathkeeper (sealed acceptance checks), The Tracker (root-cause proof), The Archivist (memory hygiene), The Lorekeeper (repo principles from history), The Inspector (QA validation).

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Workflow Automation.

    It organises the work; it does not judge code quality by itself. Done checks are as good as the commands you agree on. The board is a terminal table, not a web UI. One repo at a time.

    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 buying a skill.

    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

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed today

    Software
    GitPython 3.8+Bash

    On Windows, run it from Git Bash or WSL.

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    File Scopes

    .guild/**

    Runs git and the check commands you approve. Creates sibling git worktrees (../<repo>-<quest>) only when you ask for one. No network access.

    Listedtoday

    What's inside

    Frequently Asked Questions

    No. The script only runs git and the check commands you agree on, on your machine. It has no network code.

    No. This skill is self-contained: one Python file and plain Markdown files in .guild/. The open-source guild is a bigger tool with a local server and tmux; you do not need it.

    You choose on the first run. Shared in git so the team sees quests and rules, or private (kept out of git with .git/info/exclude).

    Any agent that can run shell commands and edit files: Claude Code, Codex, Cursor, and others. It needs git, Python 3.8+ and bash (Git Bash or WSL on Windows).

    No. It never merges, pushes or opens a PR. The wrap-up ends with your grade and your call: merge, changes or drop.

    The Quartermaster runs the whole task: goal, decisions, questions, board, wrap-up. The Oathkeeper locks the done checks with a hash and reports any test that was weakened. They work well together.

    New in Workflow Automation