plan
Generates a detailed architectural or execution plan and saves it to docs/plan/. Use this ONLY when the user explicitly asks to "plan" or "create a plan".
- License 许可
- MIT
- 最近更新
- 2026-04-09
- Stars
- ★ 246
- Forks
- 15
01 前言 Front Matter
- name
- plan
- description
- Generates a detailed architectural or execution plan and saves it to docs/plan/. Use this ONLY when the user explicitly asks to "plan" or "create a plan".
02 资源文件 3
- 参考文档references/errors.md901 B
- 参考文档references/examples.md67 B
- 参考文档references/implementation.md1.4 KB
03 SKILL.md 正文
Plan Generator
This skill captures the user's intent, formulates a comprehensive step-by-step plan, and persists it to a markdown file for documentation and future reference.
When to use this skill
- When the user says "I want to plan".
- When the user says "create a plan" or "draft a plan".
- When the user asks to "plan" a specific feature or refactor.
Instructions
- Analyze Context: Review the conversation history and the user's specific request to understand the scope (e.g., new feature, refactor, bug fix).
- Formulate Plan: Create a structured plan including:
- Objective: High-level goal.
- Context: Why we are doing this.
- Implementation Steps: Detailed breakdown of tasks (files to create, code to modify, commands to run).
- Verification: How to test the plan.
- Generate Filename: Create a descriptive filename with a timestamp to prevent overwrites.
- Format:
docs/plan/YYYY-MM-DD-topic-name.md - Example:
docs/plan/2026-02-01-nova-sonic-integration.md
- Format:
- Write File: Save the formatted plan content to the generated path.
- STOP:
- CRITICAL: Do not execute the plan.
- Do not write code (other than the plan file).
- Confirm to the user: "Plan saved to [filename]." and terminate the turn.
Constraints
- Output Location: MUST be inside
docs/plan/. - Action Limit: ONLY write the plan file. DO NOT implement the steps.
- Tone: Technical, concise, and structured.
Remember
The user request is the plan and save it to docs/plans/*.md *.md is your definition of done and stop after that.
04 同仓库更多 Skills
test
Run TTS round-trip tests — pure-logic unit tests and optional integration tests against live TTS/STT services.
speak
Speak text aloud using the Magpie TTS container in the realtime-api Docker stack. Zero external dependencies — just curl + aplay.
deploy
Build and deploy the realtime-api Docker container with full verification that deployed code matches local source.
memory
Unified memory management for notes, knowledge graph, RAG search, and file analysis. Use when working with: (1) Core memory — protected identity, projects, relationships, and system facts that should never be forgotten, (2) Working notes — per-session ephemeral notes organized by section, (3) MongoDB RAG — vector-search-enabled notes with importance scoring, decay, deduplication, and archival, (4) Neo4j knowledge graph — entities, relationships, merge duplicates, reinforce mentions, Cypher queries, (5) File analysis — deep file reading that extracts knowledge into all memory layers, (6) Service initialization — health-check, start/stop MongoDB, Neo4j, TEI embeddings via docker-compose with partial setup support.
coding
Enhanced coding assistance
memory
Unified memory management for notes, knowledge graph, RAG search, and file analysis. Use when working with: (1) Core memory — protected identity, projects, relationships, and system facts that should never be forgotten, (2) Working notes — per-session ephemeral notes organized by section, (3) MongoDB RAG — vector-search-enabled notes with importance scoring, decay, deduplication, and archival, (4) Neo4j knowledge graph — entities, relationships, merge duplicates, reinforce mentions, Cypher queries, (5) File analysis — deep file reading that extracts knowledge into all memory layers, (6) Service initialization — health-check, start/stop MongoDB, Neo4j, TEI embeddings via docker-compose with partial setup support.