Developer Tools●●Solid
Git worktrees and evidence gates for Codex and Claude Code
Git worktree isolation per task with lease-based scheduling beats naive parallelism.
Big BrainNiche Gem
alex-reyss
404d ago
Back to browse
GitHub Repository
Autonomous multi-agent orchestration engine for software repos — L0/L1/L2 agents, leases, gates, audits, git-worktree isolation. Detached dispatch by default. GPL-3.0.
22 starsShell
L0/L1/L2 agents, leases, gates, audits, Git-worktree isolation
by alexreysa·Jun 24, 2026·5 points·0 comments
AI Analysis
●●SolidWizardryBold Bet
Git-worktree isolation keeps parallel agent tasks from stepping on each other.
Strengths
- •Three-tier scheduling (L0→L1→L2) enables genuine parallel task execution.
- •State packets with JSON schemas make agent outcomes auditable and debuggable.
- •Lease system reclaims stale tasks when workers exit or timeout automatically.
Weaknesses
- •Multi-agent orchestration is becoming crowded — CrewAI, LangGraph, AutoGen exist.
- •Twenty-two stars and zero forks signals very early stage with limited adoption.
Category
Target Audience
Teams building multi-agent AI coding systems
Similar To
CrewAI · LangGraph · AutoGen
Similar Projects
AI/ML●●Solid
Agentic coding workflows built on Git worktrees and task evidence
Git worktree isolation keeps parallel AI agents from breaking main branch.
Big BrainNiche Gem
alexreysa
1023d ago
Developer Tools●●●Banger
Orc – Release the horde. Multi-agent orchestration in pure bash
Git worktrees + tmux for agent isolation beats Python orchestration frameworks.
Big BrainWizardryZero to One
mrxdev
223mo ago
Developer Tools●●●Banger
Batty – Run a team of AI coding agents in tmux with test gating
tmux-native agent orchestration with git worktrees beats CrewAI's invisible background processes.
WizardryNiche GemShip It
Zedmor
202mo ago
Developer Tools●●Solid
Fabrika, from babysitting agents to software factories
Git worktree isolation lets agents build in parallel without stepping on each other.
Big BrainShip It
berkaycubuk
2013d ago
Developer Tools●●Solid
Run approved AI repo changes from a Markdown playbook
Dual LLM audits on frozen packets with explicit stop states beat fire-and-forget agent loops.
Big BrainShip It
EagleEdge
202mo ago