WATER Self-Improving Coding Agent Model Card
这是可索引项目报告证据页：它保留 WATER Self-Improving Coding Agent 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。
WATER Self-Improving Coding Agent Model Card
| Field | Value |
|---|---|
| Repository | manthanguptaa/water |
| Category | Self-Improving Coding Agent with Benchmark-Oriented Execution |
| Stars / forks snapshot | 288 / 38 |
| Language | Python |
| License | Apache-2.0 |
| Raw capture | raw-github/manthanguptaa_water.md |
| Updated by | hourly public metadata update, 2026-06-01 19:51 +0800 |
1. Role in Self Evolve
manthanguptaa/water targets self-improving coding agents with benchmark-linked execution loops and iterative optimization behavior. It matters because self-evolving agents need repeatable harness control, measurable feedback loops, and reusable skill procedures before claiming stable improvement.
2. Working Principle
run coding-agent trajectories under measurable evaluation loops -> compare strategy revisions against task outcomes and regression signals -> retain higher-performing behaviors while dropping failing patches -> improve coding reliability through iterative self-correction
3. Evidence Path
web-observed GitHub page showed 288 stars, 38 forks, 174 commits, Apache-2.0 license, and explicit self-improving coding-agent framing. 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 Self-Improving Coding Agent with Benchmark-Oriented Execution: 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.