skills.rest
catlog22
catlog22COMMUNITY
Explore all skills by this author →
GitHub

maestro-coordinate

Coordinate multi-agent pipelines that map intents to ordered skill chains.

511|63|Updated Mar 17, 2026
Search Tags:#automation#workflow#multi-agent#orchestration#pipeline#session-management#intent-classification
One-click install
npx skills add https://github.com/catlog22/Maestro-Flow --skill maestro-coordinate
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: maestro-coordinate
Source: https://github.com/catlog22/Maestro-Flow/tree/main/.codex/skills/maestro-coordinate
Command: npx skills add https://github.com/catlog22/Maestro-Flow --skill maestro-coordinate

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Maestro Coordinate resolves ambiguous high-level intents into a deterministic, sequential chain of skill agents and manages the lifecycle of each step so large tasks are executed reliably without manual orchestration.

Core Features & Use Cases

  • Intent classification & chain resolution: Classifies user intent via keyword heuristics or an explicit chain override and maps it to a predefined ordered chain of skills.
  • Per-step agent orchestration: Spawns one agent per step, embeds the exact skill invocation in the agent prompt, waits for structured output, closes the agent, and propagates findings as context to successors.
  • Session persistence and recovery: Persists state.json per session under .workflow/.maestro-coordinate/{session-id}/ and supports --dry-run, --continue resume, auto-yes propagation, timeout handling, and failure recovery.
  • Use Case: Run end-to-end feature implementation, bug fixes, or release tasks by issuing a single intent; the coordinator will plan, execute, verify, and report results across multiple specialized agents.

Quick Start

Run maestro-coordinate with a clear intent such as implement user authentication with JWT to resolve a chain and execute the pipeline.

Frequently Asked Questions about maestro-coordinate

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I orchestrate a multi-agent pipeline for end-to-end feature implementation?▼

Multi-agent pipeline orchestration resolves a high-level intent into an ordered chain of skill agents, spawning one agent per step and propagating context sequentially. It classifies the intent, executes the mapped skill chain, and manages the lifecycle of each step automatically.

What is the best way to automate a sequential workflow for bug fixes and testing?▼

Sequential workflow automation maps a natural-language intent to a predefined ordered chain of skills for bug fixes and testing. It spawns specialized agents per step, waits for structured output, and propagates findings to successor agents for reliable execution.

Can I resume a failed multi-agent workflow session without restarting the entire pipeline?▼

Multi-agent workflow sessions support resume capability through state persistence at .workflow/.maestro-coordinate/{session-id}/state.json. You can use the --continue flag to recover from failures or timeouts without losing prior step execution context.

Does multi-agent orchestration support dry-run visibility before executing a pipeline?▼

Multi-agent orchestration supports dry-run visibility through the --dry-run flag, allowing you to preview the resolved chain of skill agents before actual execution. This helps verify intent classification and the planned sequential steps.

How does intent classification work when mapping a natural-language request to a skill chain?▼

Intent classification uses keyword heuristics or an explicit chain override to map a natural-language request to a predefined ordered chain of skills. This deterministic resolution ensures large tasks are executed reliably without manual orchestration.

What are the limitations of using multi-agent orchestration for software engineering tasks?▼

Multi-agent orchestration relies on predefined skill chains and keyword heuristics for intent classification, which may not resolve highly ambiguous intents without an explicit chain override. Timeout and failure handling are supported, but complex pipelines require careful step sequencing.

Related Skills

Explore More Software Engineering Skills→
matteoscuratimatteoscurati

8004skill

Translate natural language intents into deterministic scripts for ERC-8004 registries.

Community
Advanced
jossteijosstei
453

resume-session

Resume interrupted Maestro workflows from the first pending or failed phase.

Community
Intermediate
numman-alinumman-ali
1.0k

orchestration

Coordinate multi-agent workflows by delegating tasks to specialized agents.

Community
Advanced
Geralt1983Geralt1983
1

stream-chain

Execute multi-agent pipelines through sequential prompt chaining and predefined workflows.

Community
Advanced