Gemini CLI Auto Memory Model Card
这是可索引项目报告证据页：它保留 Gemini CLI Auto Memory 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。
Gemini CLI Auto Memory Model Card
| Field | Value |
|---|---|
| Repository | google-gemini/gemini-cli |
| Category | Agent CLI Auto-Memory and Skills |
| Stars / forks snapshot | 105000 / 13900 |
| Language | TypeScript |
| License | Apache-2.0 |
| Raw capture | raw-github/google-gemini_gemini-cli.md |
| Updated by | hourly public metadata update, 2026-05-25 16:48 +0800 |
1. Role in Self Evolve
Gemini CLI is a major open-source terminal agent; its Auto Memory docs add a mainstream implementation path for mining past sessions into reviewable durable memory and reusable Agent Skills.
2. Working Principle
session transcripts -> auto-memory mining -> reviewable patches / SKILL.md drafts -> approved durable memory or skill assets
3. Evidence Path
web-observed GitHub page showed Apache-2.0 license, 105k stars, 13.9k forks, 6,224 commits, memory-tests/evals directories, v0.43.0 latest release on 2026-05-22, and README claims for an open-source terminal AI agent; Auto Memory docs say it mines past sessions for durable memory and reusable Agent Skills, drafts patches or SKILL.md files, and never applies them without approval. 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 Agent CLI Auto-Memory and Skills in the raw -> classification -> project card -> site/report pipeline. It helps distinguish taxonomy evidence, memory substrate evidence and continuity benchmark evidence.
5. Limits
The repository was not cloned in this iteration; no benchmark, SDK example, memory experiment, eval run, skill install flow, agent loop, or production deployment was executed. Counts and claims are visible public-page/search signals unless independently revalidated later.