sd0x-dev-flow Model Card
这是可索引项目报告证据页：它保留 sd0x-dev-flow Model Card 的源材料入口、机制线索和限制提醒；正文仍需 reader/editor 与 academic public-copy review 后才能当作最终结论引用。
sd0x-dev-flow Model Card
| Field | Value |
|---|---|
| Repository | sd0xdev/sd0x-dev-flow |
| Category | Claude Code Harness Safety Runtime |
| Stars / forks snapshot | 157 / 21 |
| Language | TypeScript |
| License | MIT |
| Raw capture | raw-github/sd0xdev_sd0x-dev-flow.md |
| Updated by | hourly public metadata update, 2026-05-26 10:04 +0800 |
1. Role in Self Evolve
sd0x-dev-flow is a harness layer for Claude Code with dual-review gates, state-machine controls and fail-closed safety for coding-agent workflows. It matters because self-evolving agents need repeatable harness control, measurable feedback loops, and reusable skill procedures before claiming stable improvement.
2. Working Principle
hook lifecycle -> state-machine gates -> dual-review approvals -> fail-closed enforcement -> reusable skill pack
3. Evidence Path
web-observed GitHub page showed 157 stars, 21 forks, 456 commits, MIT license, and README claims for hook-enforced dual review, state-machine gate checks, fail-closed policies, and a large bundled public skill set for coding-agent operations. Shell GitHub API access remained blocked by DNS and local gh auth was invalid, so this card treats the snapshot as web-observed rather than API-verified.
4. Teaching Use
Use this card to explain Claude Code Harness Safety Runtime: it shows how harness/runtime/benchmark layers convert agent behavior into reproducible and auditable engineering workflows.
5. Limits
The repository was not cloned in this iteration; no benchmark run, plugin install, workflow execution, or agent loop experiment was executed. Counts and claims are visible public-page/search signals unless independently revalidated later.