S40911120
Featured
recensa
Recensa is a self-hosted web viewer that indexes Claude Code session transcripts into a local SQLite database, enabling full-text search, replay, and audit of all past agent conversations without uploading data anywhere.
fable-foreman
Fable Foreman is a Claude Code skill that turns your leading Claude model into a team lead, routing tasks to cheaper Claude subagents or OpenAI Codex workers based on capability classes with blind verification.
Stars
93
7-day growth
No data
Forks
26
Open issues
0
License
MIT
Last updated
2026-07-19
AI repository intelligence
FR-AI / ANALYSISWhy it is worth attention
It implements a multi-agent system with cost discipline and verification that addresses common runaway cost problems, using runtime model probing and capability-class routing that never needs updating for new models.
Who it is for
- Claude Code users managing complex coding tasks
- AI engineers seeking cost-effective multi-agent workflows
- Developers wanting automated verification of delegated work
Use cases
- Multi-file or multi-stage coding projects
- Task delegation with quality verification at lower cost
- Cost-controlled parallel execution using a mix of Claude and OpenAI Codex
- Safe automated refactoring with blind verification of changes
Strengths
- Routes by capability class (frontier, workhorse, fast) not model IDs, so new model releases require zero skill updates
- Blind verifier with fresh context cross-checks worker output, preventing overconfidence
- Respects budget with sequential dispatch, announced fan-outs, and visible degradation under quota pressure without lowering quality bar
- No configuration files or enforcement scripts—one skill, three agent roles, and rules a frontier model follows
Considerations
- Full orchestration requires Claude Code and proper installation of both skill and agent files; claude.ai/Desktop runs only weaker discipline mode
- Windows plugin installation is broken due to a Claude Code bug (EPERM, not planned to fix); use paste-in method instead
- Single conversation model same-model self-review (discipline mode) is weaker than cross-model blind verification
README quick start
Fable Foreman
Your strongest model shouldn't be swinging the hammer.
Built in public by DontSleepOnAI — the story behind this skill (including the five-round adversarial review where OpenAI's newest model tore apart the first draft) lives there.
Fable Foreman turns whichever frontier-class Claude model leads your session into a team lead: it plans, routes each task to the cheapest worker that clears the quality bar — Claude subagents or OpenAI Codex CLI workers, auto-detected — and, in full orchestration mode, refuses to accept meaningful changes until a blind, fresh-context verifier reproduces the evidence. (Environments without subagents get an honest reduced-assurance mode that says so.)
No dated model IDs in routing policy. No configuration files. No enforcement scripts. One skill, three agent roles, and a set of rules good enough that a frontier model actually follows them.
"Fable" is where it started, not what it needs. The foreman seat is a capability class, so any frontier-class Claude runs the skill identically — Opus leads it exactly as Fable does, with the same routing tree, gates, and verification contract. That holds whether an Opus session invokes the skill directly or a Fable session falls back to Opus mid-run; the skill re-probes its own seat and carries on rather than routing off a stale identity.
Why
Anthropic's own engineering shows both sides of the ledger. Their multi-agent research system writeup found an orchestrator-plus-cheaper-subagents design strongly outperformed single agents — an Opus lead with Sonnet workers, exactly this skill's shape — while consuming roughly 15x the tokens of a single chat, which is why they conclude multi-agent work only pays for high-value tasks. Anthropic's own cost guidance likewise recommends cheaper-tier teammates under a stronger lead as the default for multi-agent work. And the community has receipts for what happens without discipline — runaway-subagent cost stories are a genre of their own on every AI-coding forum, which is exactly why this skill bounds crew sizes, retries, and spend announcements the way it does.
The difference between those two outcomes is not orchestration machinery — it's **routing judgment and verification disciplin
Description
Turn your strongest Claude model into a team-lead orchestrator: it plans and reviews while cheaper Claude or Codex workers execute — with blind verification before anything counts as done.
Related repositories
Similar projects matched by category, topics, and programming language.
bjan
Featured
claudemux
Claudemux is a lightweight, dependency-minimal message bus that lets multiple Claude Code sessions communicate in real-time over tmux, enabling one session to ask another a question and receive the answer without manual intervention.
Vincentwei1021
video-shotcraft
An AI agent skill that turns Claude Code or Codex into a motion-design studio for crafting cinematic product videos with Remotion, offering 106 shot recipe cards, 162 styles, 161 motion previews, and a production-ready template.