More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Buer Pro Max

    1

    Bootstrap a cross-session project governance framework with plans, ID sequences, and authorization state machines.

    $6

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    buer-pro-max

    buer-pro-max

    Example session with this skill installed

    My project is getting messy and I keep forgetting what I planned in previous sessions. Bootstrap a governance framework for this repo.

    • Read your context and instructions
    • Compiled the buer-pro-max

    I have completed the due diligence for this repository. Based on the 12,000 lines of Python code, the existence of a Git VCS, and the presence of external upstream dependencies in vendor/, I have selected the

    Standard Tier for your governance framework.

    Governance Framework Initialized

    The governance root has been established at docs/governance/. I have also appended the Project Constitution to your AGENTS.md to ensure these rules are respected in all future sessions.

    Key Components Installed:

    • Active Plans Container: Located at docs/governance/活跃计划/ for current work.
    • Issue & Deviation Ledgers: Numbered sequences for tracking ISS and DEV items.
    • Authorization State Machine: Rules defining the "Register -> Dispatch -> Execute" flow.
    • Closure Ritual: A 5-step verification process required before any plan is marked as complete.

    Directory Structure:

    docs/governance/
    ├── README.md               <-- Start here for reading order
    ├── 工作范式.md              <-- The authoritative rulebook
    ├── 项目清单/                <-- Due diligence and scope maps
    ├── 计划队列/                <-- Pending items
    ├── 活跃计划/                <-- Current session tasks
    ├── 归档计划/                <-- Historical records
    ├── 问题记录/                <-- ISS-001, ISS-002...
    └── 代码修改记录/             <-- DEV-001, DEV-002...
    

    Next steps

    1. Read docs/governance/README.md to understand the default session reading order.
    2. Review the Writable Scope in 工作范式.md to confirm the vendor/ directory is correctly locked.
    3. Start your first plan by creating a new file in 计划队列/ using the _队列项模板.md.

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

    About this skill

    超过 5,000 行或跨越多个会话的项目,常常会因上下文丢失、未记录的偏离和范围蔓延而不堪重负。本技能强制执行一套跨会话治理框架,通过确保未经授权不得执行任何操作来防止返工。

    它做什么

    • 引导式治理 :创建结构化目录树,用于跟踪计划、决策和问题,并根据项目规模进行调整。

    • 强制授权:实现“登记、派发、执行”状态机,其中口头“同意继续”不算作许可。

    • 维护证据:强制要求验证绝不能由执行代码的人本人完成。

    • 定义收尾:用严格的多步骤仪式取代含糊的“完成了”说法,包括受门禁检查的检查和台账更新。

    • 上下文恢复:生成交接提示词,确保后续 AI 会话理解当前状态和决策顺序。

    它如何工作

    • 尽职调查:扫描仓库中的版本控制系统、上游依赖和构建工具,以确定必要约束。

    • 层级选择:根据代码量和会话历史,分配五种配置之一(从微型到重型)。

    • 脚手架搭建:生成治理根目录,并将一份“宪章”注入 AGENTS.md 或 CLAUDE.md。

    • 主动跟踪:管理计划从“活跃”到“归档”的生命周期,并为问题和偏离使用唯一 ID 序列。

    框架与工具

    这是一个与宿主无关、基于 Markdown 的框架。它使用 POSIX shell 命令进行仓库分析,并适用于任何技术栈,包括 React、Python、Rust 或 monorepo。它针对在 Claude Code 或 Cursor 中使用进行了优化。

    为什么这胜过自己写提示词

    通用提示词无法在长历史中维持严格的 ID 序列,也无法防止“幻觉式授权”。本技能提供了逻辑基础设施,使项目在数百个会话中保持稳定,而不丢失任何一项决策。

    使用场景

    • 稳定一个 50k+ 行的个人项目,而你经常忘记某些架构选择为何做出。

    • 让子代理接入一个遗留仓库,并对上游代码设置严格的“只读”边界。

    • 将项目从松散的 README 笔记转变为专业、可供审计的合规台账。

    已知限制

    不适用于小型脚本或周末实验,因为治理开销超过返工成本。初始尽职调查探测需要兼容 POSIX 的 shell。

    需要注意

    使用本skills因为涉及到项目治理,会频繁读取写入治理文件,token消耗会有提升。

    ------以下为英文原文。 (The following is the original English text.)------

    Projects exceeding 5,000 lines or spanning multiple sessions often collapse under the weight of context loss, undocumented deviations, and scope creep. This skill enforces a cross-session governance framework that prevents rework by ensuring no execution happens without authorization.

    What it does

    • Bootstrap governance creates a structured directory tree for tracking plans, decisions, and issues tailored to the project scale.
    • Enforce authorization implements a register, dispatch, and execute state machine where verbal "go aheads" do not count as permission.
    • Maintain evidence mandates that validation is never performed by the person who executed the code.
    • Define closure replaces vague "done" statements with a rigorous multi-step ritual involving gated checks and ledger updates.
    • Context recovery generates handoff prompts to ensure subsequent AI sessions understand the current state and sequence of decisions.

    How it works

    1. Due diligence scans the repo for VCS, upstream dependencies, and build tools to determine the necessary constraints.
    2. Tier selection assigns one of five profiles (Micro to Heavy) based on code volume and session history.
    3. Scaffolding generates the governance root directory and injects a "constitution" into AGENTS.md or CLAUDE.md.
    4. Active tracking manages the lifecycle of plans from "Active" to "Archived" with unique ID sequences for issues and deviations.

    Frameworks & tools

    This is a host-agnostic Markdown-based framework. It uses POSIX shell commands for repo analysis and works with any stack, including React, Python, Rust, or monorepos. It is optimized for use within Claude Code or Cursor.

    Why this beats prompting it yourself

    Generic prompts fail to maintain strict ID sequences or prevent "hallucinated authorization" over long histories. This skill provides the logical infrastructure to keep a project stable across hundreds of sessions without losing a single decision.

    Use cases

    • Stabilizing a 50k+ line solo project where you frequently forget why certain architectural choices were made.
    • Onboarding a sub-agent to a legacy repo with strict "read-only" boundaries for upstream code.
    • Moving a project from loose README notes to a professional audit-ready ledger for compliance.

    Known limitations

    Not suitable for small scripts or weekend experiments where the overhead of governance exceeds the cost of rework. Requires a POSIX-compliant shell for initial due diligence probes.

    Please note

    Because using this skill involves project governance, governance files will be read and written frequently, which will increase token consumption.

    Requires POSIX shell (bash/zsh/git-bash). Not for one-off scripts or projects with <5k lines. Python 3 needed for self-check scripts.

    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

    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 3 days ago

    Requires

    Git
    markdown
    Python
    Windows only

    Permissions

    Terminal / Shell
    Network Access
    Read Files
    Write Files
    Listed3 days ago

    What's inside

    Frequently Asked Questions

    Bootstrap a cross-session project governance framework with plans, ID sequences, and authorization state machines.

    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 buer-pro-max 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 Agents & Orchestration