weave

Orchestrate AI coding workflows across intake, context, execute, and close phases.

Updated Feb 16, 2026
One-click install
npx skills add https://github.com/AGM1968/weave --skill weave
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: weave
Source: https://github.com/AGM1968/weave/tree/main/.claude/skills/weave
Command: npx skills add https://github.com/AGM1968/weave --skill weave

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Weave provides a graph-first orchestrator to manage end-to-end AI coding workflows across intake, context, execute, and close phases, ensuring consistent graph semantics and learnings capture.

Core Features & Use Cases

  • Graph-first workflow orchestration for AI coding agents in software repositories.
  • Context Pack generation, blocker validation, and learning capture across tasks such as epics and tasks.
  • Lightweight integration with optional resources (scripts, references, assets) and routing rules for tiered model usage.

Quick Start

Run weave to initialize and load the INTAKE phase for your repository.

Frequently Asked Questions about weave

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

FAQPage Schema
How do I orchestrate AI coding workflows with a task graph?

You can orchestrate AI coding workflows using a graph-first orchestrator like weave, which coordinates intake, context, execute, and close phases. It structures software repository tasks into a task graph, managing blockers and capturing learnings across epic and task hierarchies.

What is graph-first orchestration for AI coding agents?

Graph-first orchestration coordinates AI coding agents by enforcing a four-phase lifecycle across software repositories. It validates blockers, generates context packs, and provides routing, caching, and recovery capabilities to maintain consistent graph semantics throughout the workflow.

How do I generate context packs for AI coding tasks?

To generate context packs for AI coding tasks, run a graph-first orchestrator to load the INTAKE phase for your repository. The orchestrator automatically handles context pack generation and blocker validation throughout the structured task graph execution.

Can I use weave for software-repo projects requiring structured task graphs?

Yes, weave is designed specifically for software-repo projects requiring structured task graphs. It supports epic and task hierarchies, applies optional resources like scripts and references, and enforces a four-phase lifecycle with routing rules for tiered model usage.

How do I capture learnings across epic and task hierarchies in AI workflows?

You capture learnings across epic and task hierarchies using a graph-first orchestrator that enforces a close phase. This phase finalizes the workflow, ensuring consistent graph semantics and learning capture across the entire software repository project lifecycle.

Does weave support routing rules for tiered model usage in AI workflows?

Yes, weave supports routing rules for tiered model usage in AI workflows. It provides routing, caching, and recovery capabilities to manage AI coding agents effectively, ensuring consistent graph semantics across the intake, context, execute, and close phases.