HN 표시: Fabro – 오픈 소스 다크 소프트웨어 공장

AI coding agents are powerful but unpredictable. You either babysit every step or review a 50-file diff you don't trust. Fabro gives you a middle path: define the process as a graph, let agents execute it, and intervene only where it matters. Why Fabro? # With Claude Code curl -fsSL https://fabro.sh/install.md | claude # With Codex codex "$(curl -fsSL https://fabro.sh/install.md)" # With Bash curl -fsSL https://fabro.sh/install.sh | bash - Extend disengagement time — Stop babysitting an agent REPL. Define a workflow with verification gates and walk away — Fabro keeps the process on track without you. - Leverage ensemble intelligence — Seamlessly combine models from different vendors. Use one model to implement, another to cross-critique, and a third to summarize — all in a single workflow. - Share best practices across your team — Collaborate on version-controlled workflows that encode your software processes as code. Review, iterate, and reuse them like any other source file. - Reduce token bills — Route cheap tasks to fast, inexpensive models and reserve frontier models for the steps that need them. CSS-like stylesheets make this a one-line change. - Improve agent security — Run agents in cloud sandboxes with full network and filesystem isolation. Keep untrusted code off your laptop and out of your production environment. - Run agents 24/7 — Fabro's API server queues and executes runs continuously. Close your laptop — workflows keep running and results are waiting when you return. - Scale infinitely — Move execution off your laptop and into cloud sandboxes. Run as many concurrent workflows as your infrastructure allows. - Guarantee code quality — Layer deterministic verifications — test suites, linters, type checkers, LLM-as-judge — into your workflow graph. Failures trigger fix loops automatically. - Achieve compounding engineering — Automatic retrospectives after every run feed a continuous improvement loop. Your workflows get better over time, not just your code. - Specify in natural language — Define requirements as natural-language specs and let Fabro generate — and regenerate — implementations that conform to them. | Feature | Description | | |---|---|---| | 🔀 | Deterministic workflow graphs | Define pipelines in Graphviz DOT with branching, loops, parallelism, and human gates. Diffable, reviewable, version-controlled | | 🙋 | Human-in-the-loop | Approval gates pause for human decisions. Steer running agents mid-turn. Interview steps collect structured input | | 🎨 | Multi-model routing | CSS-like stylesheets route each node to the right model and provider, with automatic fallback chains | | ☁️ | Cloud sandboxes | Run agents in isolated Daytona cloud VMs with snapshot-based setup, network controls, and automatic cleanup | | 🔌 | SSH access and preview links | Shell into running sandboxes with fabro sandbox ssh and expose ports with fabro sandbox preview for live debugging | | 🌲 | Git checkpointing | Every stage commits code changes and execution metadata to Git branches. Resume, revert, or trace any change | | 📊 | Automatic retros | Each run generates a retrospective with cost, duration, files touched, and an LLM-written narrative | | ⚡ | Comprehensive API | REST API with SSE event streaming and a React web UI. Run workflows programmatically or as a service | | 🦀 | Single binary, no runtime | One compiled Rust executable with zero dependencies. No Python, no Node, no Docker required | | ⚖️ | Open source (MIT) | Full source code, no vendor lock-in. Self-host, fork, or extend to fit your workflow | A plan-approve-implement workflow where a human reviews the plan before the agent writes code: digraph PlanImplement { graph [ goal="Plan, approve, implement, and simplify a change" model_stylesheet=" * { model: claude-haiku-4-5; reasoning_effort: low; } .coding { model: claude-sonnet-4-5; reasoning_effort: high; } " ] start [shape=Mdiamond, label="Start"] exit [shape=Msquare, label="Exit"] plan [label="Plan", prompt="Analyze the goal and codebase. Write a step-by-step plan.", reasoning_effort="high"] approve [shape=hexagon, label="Approve Plan"] implement [label="Implement", class="coding", prompt="Read plan.md and implement every step."] simplify [label="Simplify", class="coding", prompt="Review the changes for clarity and correctness."] start -> plan -> approve approve -> implement [label="[A] Approve"] approve -> plan [label="[R] Revise"] implement -> simplify -> exit } Agents run as multi-turn LLM sessions with tool access. Human gates (hexagon ) pause for approval. The stylesheet routes planning to a cheap model and coding to a frontier model. See the Graphviz DOT language reference for the full syntax. Fabro ships with comprehensive documentation covering every feature in depth: - Getting Started -- Installation, first workflow, and why Fabro exists - Defining Workflows -- Node types, transitions, variables, stylesheets, and human gates - Executing Workflows -- Run configuration, sandboxes, checkpoints,