agentmemory Model Card
这是可索引项目报告证据页：它保留 agentmemory 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。
agentmemory Model Card
One Sentence
agentmemory is the cross-harness memory layer that turns continuity from a promise into an engineering dependency.
Three Sentences
It targets Claude Code, Codex, Cursor, Hermes, OpenClaw, and MCP clients instead of staying inside one runtime. That makes it a direct answer to the user’s memory-substrate requirement for self-evolving agents. The 2026-06-17 morning packet keeps the memory-substrate evidence current with another small but real GitHub delta.
Model Card
| Field | Value |
|---|---|
| Repository | rohitg00/agentmemory |
| Source | raw-github/rohitg00_agentmemory.md |
| Category | Agent persistent memory layer |
| Pattern | persistent memory -> wiki-style knowledge capture -> graph/hybrid retrieval -> cross-harness continuity |
| Evidence | Authenticated GitHub API snapshot, 2026-06-17 08:29 +0800 |
Teaching Use
Use agentmemory to teach why memory cannot be an afterthought. Durable facts, project traces, user preferences, and retrieval boundaries are what let an agent accumulate value across sessions.
Evidence And Limits
The raw capture now reflects a GitHub metadata packet observed on 2026-07-05: 24,554 stars, 2,021 forks, 463 commits, 172 open issues, and 188 open pull requests. This packet is fresher than the previous authenticated packet at 2026-07-05 01:38 +0800 where a delta was observed. This run did not execute the repository locally, validate workflows end to end, or independently rerun benchmark claims. Product, memory, benchmark, and automation claims therefore remain repository-scoped unless separately tested.