hermes2anti Model Card
这是可索引项目报告证据页：它保留 hermes2anti 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。
hermes2anti Model Card
| Field | Value |
|---|---|
| Repository | swapedoc/hermes2anti |
| Category | Memory and Skill Self-Improvement Toolkit |
| Stars / forks snapshot | 4 / 0 |
| Language | Python |
| License | MIT |
| Raw capture | raw-github/swapedoc_hermes2anti.md |
| Updated by | hourly public metadata update, 2026-05-26 05:44 +0800 |
1. Role in Self Evolve
hermes2anti is a small implementation of the memory-skill loop popularized by Hermes-style agents: it persists lessons, creates reusable skills, scans those skills, and recalls prior sessions.
2. Working Principle
session starts -> load memory and skills -> complete task -> extract golden path -> create and scan skill -> save memory -> next session starts smarter
3. Evidence Path
Web-observed GitHub evidence showed 4 stars, 0 forks, 1 commit, MIT license, and a Python-only implementation. README components include memory management, skill management, security scanning, SQLite full-text recall, and git-history insights.
4. Teaching Use
Use this card as a minimal example of how self-improvement can be decomposed into memory, reusable procedures, safety scanning, and retrieval.
5. Limits
The repository appears early-stage, and no local execution or security scan was run. Shell GitHub API DNS failed, so metadata is web-observed rather than API-verified.