Experimental fork — ingesting features from claw-code (Rust) into claw-code-go (Go)
Warning: This is an ongoing experiment and is not tested. Use at your own risk.
This repo is a fork of daolmedo/claw-code-go, a Go reimplementation of Claude Code — Anthropic's agentic coding assistant.
On top of the upstream codebase, this fork ingests features from ultraworkers/claw-code — a Rust port of Claude Code that emerged after the source leak. The goal is to bring the more complete Rust implementation's capabilities into the Go codebase: hooks, plugins, MCP lifecycle, sandbox, permissions engine, runtime session management, apikit with multi-provider routing, prompt caching, and much more.
This is an experiment, not a product. The ported code has not been manually tested or validated beyond automated build checks.
The Rust-to-Go feature porting was orchestrated by Iterion — a workflow engine for complex multi-agent LLM pipelines, using Claude Code, Codex, and other backends.
This repo serves as the real-world example that drove Iterion's development. The full workflow configuration and writeup are available:
- rust_to_go_port.iter — the Iterion workflow definition (DSL)
- rust_to_go_port.md — design decisions, lessons learned, and empirical data
| Metric | Value |
|---|---|
| Refinement runs | 4 (workflow iterated alongside the engine) |
| Outer loop iterations | 25+ batches across all runs |
| Commits added | 48 (47 by Iterion, 1 by Claude Code) |
| Go code generated | 37,995 lines across 173 files + 96 test files |
| New Go packages | 30+ (hooks, plugin, apikit, worker, lane, policy, recovery, sandbox, lsp, task, team...) |
| Feature parity | 100% (37/37 features) |
| Final run | 100% parity in 41 min, single batch, zero fix loops, zero interventions |
| Longest autonomous stretch | 2h25m without human intervention |
| Dual-judge verdicts | 30+ across all runs (Claude + Codex) |
The workflow breaks the porting into dependency-ordered batches. Each batch goes through: plan → implement → simplify → commit → test → parity scan → dual-judge review → fix loop. Session continuity (fork/inherit) preserves KV cache across related phases. A human gate pauses for high-risk batches and auto-approves routine ones.
See the full writeup for details on convergence strategies, stagnation detection, and model allocation.
Experimental. The code compiles and passes go vet, but has not been manually tested. The latest commit (Anthropic prompt caching with cache_control breakpoints) was done directly with Claude Code outside the Iterion workflow.
Contributions, testing, and feedback are welcome.
- daolmedo/claw-code-go — the original Go port of Claude Code (upstream)
- ultraworkers/claw-code — the Rust port of Claude Code (feature source)
- Iterion — the workflow orchestration engine that performed the porting
MIT