fable-team
CommunityOrchestrate multi-agent teams with enforced guardrails
Software Engineering#multi-agent#orchestration#worktree#claude-code#guardrails#context-management#team-harness
Authoryoonhwan
Version1.0.0
Installs0
System Documentation
What problem does it solve?
This skill eliminates the chaos of uncoordinated multi-agent AI workflows by enforcing strict role separation between orchestrator, planner, and workers, while implementing physical guardrails against context explosion and model leaks.
Core Features & Use Cases
- Strict Role Separation: Enforces distinct models and effort levels for orchestrator (opus-4-8), planner (opus-4-8/fable-5), implementer (opus-4-6), tester (sonnet-5), checker (sonnet-4-6), and DA (codex gpt-5.5).
- Context Safety: Implements write-through state externalization to
.fable-team/state/, 300k/450k token hard gates, and automatic hygiene to prevent context loss during long-running pipelines. - Integration Ready: Supports baton (worktree navigation) and cairn (project ledger) integrations with configurable headless override policies for CI/CD environments.
Quick Start
Use the fable-team skill to orchestrate a new feature by saying "FT 구성" or "fable-team" and follow the guided boot sequence to configure your team and execute the pipeline with enforced guardrails.
Dependency Matrix
Required Modules
None requiredComponents
references
💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: fable-team Download link: https://github.com/yoonhwan/ai-feature-pack/archive/main.zip#fable-team Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 536,000+ vetted skills library on demand.