Plan Orchestrator logo

Plan Orchestrator

Organization
SilentLake-Tech-SkillHub
plan-orchestrator

Create and maintain an execution Plan after product and technical reviews approve a build, or for an explicitly scoped design or demo before product review. Do not use during ordinary requirement clarification or review.

Overview

PublisherSilentLake-Tech-SkillHub
RepositorySilentAgentHub
Skill nameplan-orchestrator
Stars
184
Forks
0
Bundled files
1
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by SilentLake-Tech-SkillHub on GitHub. Read the source before you install it.

Installation

Install the Plan Orchestrator AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/SilentLake-Tech-SkillHub/SilentAgentHub/tree/main/harness/V5.2.5/skills-source/plan-orchestrator
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/SilentLake-Tech-SkillHub/SilentAgentHub.git /tmp/SilentAgentHub
mkdir -p .claude/skills
cp -r /tmp/SilentAgentHub/harness/V5.2.5/skills-source/plan-orchestrator .claude/skills/plan-orchestrator
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Plan Orchestrator in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Plan Orchestrator on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Plan Orchestrator is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Plan Orchestrator

Mandatory gate

First identify the lifecycle stage. Create a Plan only (1) after requirement review and technical review have approved a solution and before its build, or (2) after requirement clarification, before requirement review, when the user has chosen a bounded design or Demo to make. Complexity determines whether a build in either window needs a Plan; it does not advance the lifecycle stage. Requests to discuss a solution, perform research, or review requirements/technology do not trigger a Plan. A pre-review design/Demo Plan authorizes only that artifact, not full product implementation.

For every complex construction task in an allowed Plan window, create:

text
流程管理/执行计划/<任务ID>_<任务名>/plans.md

The Plan is the single source of truth for execution. Use the state sequence draft -> awaiting_review -> approved -> executing -> validating -> awaiting_acceptance -> closed. Do not begin engineering before approved; a generic instruction such as “直接执行” does not bypass review for complex work.

When a Plan is first created or materially changes while awaiting_review, classify the current user Query as plan_create_or_update and register the exact conversation/turn and Plan path through .codex/hooks/plan-disclosure-runtime.ps1 -Event Register. If the user explicitly asks to see the complete Plan again, register plan_redisclose. Approval, rejection without a Plan edit, status questions, ordinary follow-ups, and unrelated Queries must not register a disclosure.

The same registered turn's final response must reproduce the complete current plans.md body and explicitly request approval. A path, link, summary, excerpt, or statement that the Plan exists is insufficient. Stop checks only the Plan explicitly bound to the current Query/turn; it must never scan global awaiting_review files or choose by modification time. The first Stop attempt reinjects that bound Plan once. If Memory also needs closeout, complete it in that same continuation so the next Stop can persist the disclosure receipt and display the Plan response.

Required content

  • Purpose: user result and business value.
  • Scope: included, excluded, dependencies, modules, and delivery stage.
  • Context: current facts, relevant paths, rules, comments, and constraints.
  • Plan of Work: staged actions, ownership, order, migration, and rollback.
  • Validation: checks, acceptance evidence, and substitutes for unavailable validation.
  • Progress: timestamped state and completed, active, pending, or blocked items.
  • Decision Log: options, chosen direction, reason, and user approval.
  • Surprises & Discoveries: new facts, issues, risks, and resulting plan changes.
  • Outcomes & Retrospective: delivered result, unverified or deferred work, acceptance, and lessons.

Maintenance

Update Progress after each stage, Decision Log after each material choice, and Surprises immediately when execution differs from plan. Reconcile the Plan with task, requirement-status, issue, risk, acceptance, version, and handoff records. A partially completed Plan cannot be reported as final completion.

Technical-design handoff

For the post-review build window, consume the approved technical design and recorded review decision produced via technical-design-authoring. The Plan governs engineering execution; requirements and the approved technical design govern product scope and implementation contracts. Return an unresolved technical design/review to that Skill before planning construction; preserve the existing bounded design/Demo exception.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Plan Orchestrator AI skill do?

Create and maintain an execution Plan after product and technical reviews approve a build, or for an explicitly scoped design or demo before product review. Do not use during ordinary requirement clarification or review.

Why use Plan Orchestrator on TypingMind?

Because you install it once and use it with any model. Plan Orchestrator is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Plan Orchestrator in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/SilentLake-Tech-SkillHub/SilentAgentHub/tree/main/harness/V5.2.5/skills-source/plan-orchestrator. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Plan Orchestrator?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Plan Orchestrator?

As many as you like. As long as a model supports skills, you can use Plan Orchestrator with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Plan Orchestrator AI skill free?

It is published on GitHub by SilentLake-Tech-SkillHub. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇