davidabeyer

davidabeyer / formaltask

Public

Declarative orchestration for parallel Claude Code agents — define quality gates in YAML, enforce them automatically

13
1
100% credibility
Found Feb 18, 2026 at 13 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

FormalTask is an open-source declarative orchestration tool for parallel Claude Code agents that defines quality gates in YAML and enforces them automatically using isolated git worktrees, task dependencies, and automated reviews.

Star Growth

See how this repo grew from 13 to 13 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is formaltask?

FormalTask is a Python CLI for structured task management in AI-assisted dev workflows, optimized for parallel Claude agents. It decomposes epics into formal tasks with YAML specs, spawns isolated git worktree workers, and enforces quality gates like reviews, tests, and auto-fixer spawns. Run `ft epic-decompose my-epic` to load tasks into SQLite, then `ft work spawn` or `ft work dashboard` for TUI monitoring—structured tasks that scale to 10 agents without chaos.

Why is it gaining traction?

Its rules-based kernel lets you override completion checks or orchestration via simple YAML—no recoding agents—while parallel tmux workers handle git branches and PRs automatically. Unlike loose prompt chaining, it blocks merges until reviews pass and nudges stalled tasks, delivering github llm structured output you can trust. Devs dig the inbox for blocked workers and auto-spawn for CI fails.

Who should use this?

Claude Code power users tackling feature epics, solo devs scaling AI for backend services, or teams in structured RAG pipelines needing task scopes with deps. Ideal if you're spawning agents for codegen but hate manual coordination.

Verdict

Promising for Claude-heavy workflows, with crisp CLI and TUI, but 13 stars and 1.0% credibility scream early days—test coverage looks solid yet expect rough edges. Prototype it for personal projects; hold for teams.

(187 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.