state-trace Agent Memory Engine Model Card

这是可索引项目报告证据页：它保留 state-trace Agent Memory Engine 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。

state-trace Agent Memory Engine Model Card

1. Role in Self Evolve

state-trace is a graph-native memory engine for AI agents that stores typed memories and causal links, then retrieves state-aware context through graph traversal instead of only vector similarity.

2. Working Principle

agent log step -> typed memory node/edge -> capacity-aware decay -> graph traversal retrieval

3. Evidence Path

web-observed GitHub page showed 3 commits, Python package layout, 1 star, 0 forks, and README language for graph-native AI agent memory with typed nodes, causal links, capacity-aware decay/compression/removal, JSON persistence and optional embeddings; shell GitHub API freshness was blocked.

4. Teaching Use

Use this card to explain why agent memory needs provenance, retrieval gates and degradation/forgetting rules before it can become a durable self-evolution substrate.

5. Limits

The repository was not cloned in this iteration; no local install, benchmark rerun, security review, skill execution or CI gate was performed. Counts and claims are visible public-page signals unless independently revalidated later.