← 返回 Skills 市场

落地鸿沟补全器

Name: 落地鸿沟补全器
Author: beachanger

作者 BeAChanger · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

221

总下载

当前安装

版本数

在 OpenClaw 中安装

/install blueprint-to-deployment

功能描述

将 Agent 规划结果补全为可交付、可部署的落地闭环。适用于把蓝图、架构、next actions 写入目标 Agent workspace，并主动推进到平台部署与接入引导。

使用说明 (SKILL.md)

落地鸿沟补全器

用途

这只 skill 用来补全“从规划到可用 Agent”之间的鸿沟。

很多 Agent 流程会停在：

聊天里
方案里
蓝图里
看起来已经想清楚了

但没有真正完成：

文档落地
workspace 归档
用户交付说明
部署平台追问
平台接入引导

本 skill 的目标，就是把这些缺口补上。

什么时候用

当你已经有：

Agent 蓝图
架构规划
MVP 边界
next actions

但还没真正把它交到“目标 Agent 自己的 workspace”里，或还没推进到平台接入时，触发本 skill。

核心原则

每只 Agent 都应拥有自己的资产
母 skill 保留方法论，目标 Agent 保留自己的交付资产
文档归属也是上下文管理
设计不能停留在聊天里
如果目标是可用，就必须主动推进到部署层

标准流程

第 1 步：确认目标 Agent workspace

先确认目标 Agent 的 workspace 已存在，或即将被创建。

第 2 步：写入关键交付文档

默认至少写入：

蓝图文档
架构 / MVP / 边界说明
next actions
当天推进记录

推荐落位：

memory/reports/
memory/cards/
memory/YYYY-MM-DD.md
skills/

第 3 步：给用户交付地图

明确告诉用户：

workspace 路径
文件路径
每个文件的作用
后续怎么改、去哪里看

第 4 步：主动问部署平台

不要等用户自己问。主动追问：

Telegram
Feishu
Discord
其他 channel

第 5 步：按平台继续引导

根据用户选择的平台，说明：

需要什么 bot / app / account
需要哪些配置
如何做首次联调

为什么值得单独做成 skill

因为这不是某一只母虾私有的补丁，而是所有“会产出蓝图”的 Agent / skill 都可能需要的后半段闭环能力。

它解决的问题不是“如何规划”，而是： 如何避免规划停在纸面上。

输出要求

至少交付：

目标 workspace 已写入哪些内容
这些文件分别在哪
下一步该选哪个平台
平台接入该怎么开始

反模式

只给聊天建议，不写文件
文档全留在主 Agent / 母 skill workspace
不告诉用户文件在哪里
不主动问平台
让流程停在“我们已经设计完了”

安全使用建议

This skill's purpose (turn plans into deliverables and guide deployment) is reasonable, but check two things before installing/using it: 1) Where do the recommended paths map to in your environment? Ensure the agent is only allowed to write non-executable documentation (avoid letting it create/modify files under `skills/` or other directories that contain executable code). Require explicit user confirmation before any write to sensitive paths. 2) Expect the skill to ask for platform credentials or bot tokens during runs; decide where those secrets will be entered and how they'll be stored/audited. If you can't confirm the mapping of workspace paths or you don't want the agent to modify skill code, treat this skill as risky and restrict its write permissions or have it output proposed content for manual review instead.

功能分析

Type: OpenClaw Skill Name: blueprint-to-deployment Version: 1.0.0 The skill bundle consists of metadata and markdown instructions (SKILL.md) designed to automate the transition from planning to deployment. It instructs the AI agent to write documentation, blueprints, and action plans to specific workspace directories and guides the user through platform-specific setup (e.g., Telegram, Discord). No executable code, data exfiltration logic, or malicious prompt injections were found; the instructions are functionally aligned with the stated purpose of deployment orchestration.

能力评估

ℹ Purpose & Capability

The name/description (从规划到落地的闭环) aligns with the instructions to write blueprints, next actions, and deployment guidance into a target Agent workspace. Recommending memory/paths such as memory/reports/ and memory/YYYY-MM-DD.md is proportionate. However, the guidance also recommends writing into `skills/`, which is typically where executable skill code lives — that scope is broader than 'documentation/delivery' and deserves justification.

⚠ Instruction Scope

SKILL.md explicitly instructs the agent to create and write files into a target workspace and to '主动推进' to deployment platforms (Telegram, Feishu, Discord). Writing documentation is expected, but targeting `skills/` or adding files that could be treated as skills or code is scope creep and could modify agent behavior. The instructions do not constrain what may be written (e.g., strictly non-executable docs), nor do they require user confirmation before modifying sensitive paths. The doc also implies the agent may request platform/account info during the run — the skill does not declare how such secrets are handled.

✓ Install Mechanism

No install specification and no code files — instruction-only. This is the lowest install risk (nothing is fetched or executed from external URLs).

ℹ Credentials

The skill declares no required environment variables or credentials, which is coherent for a documentation/handoff tool. However, the runtime guidance expects the agent to initiate platform onboarding and ask about bots/apps/accounts; that may lead to requests for tokens/credentials at runtime even though none are declared. The absence of declared credential requirements is plausible but the skill should explicitly state how credentials will be requested, used, and stored.

⚠ Persistence & Privilege

always:false and normal autonomous invocation are appropriate. The concern is that instructions encourage writing into `skills/` (potentially modifying executable skill code) and target other agents' workspaces; that is effectively granting the skill write access beyond passive documentation. The skill does not require or document user confirmation flows, nor does it limit writes to non-executable docs.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install blueprint-to-deployment
安装完成后，直接呼叫该 Skill 的名称或使用 /blueprint-to-deployment 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release: bridge planning to delivery by writing docs into target agent workspace, showing handoff map, and proactively guiding platform deployment.

元数据

Slug blueprint-to-deployment

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题