docs.rs failed to build rsclaw-memory-0.1.0
Please check the build logs for more information.
See Builds for ideas on how to fix a failed build, or Metadata for how to configure docs.rs builds.
If you believe this is docs.rs' fault, open an issue.
Please check the build logs for more information.
See Builds for ideas on how to fix a failed build, or Metadata for how to configure docs.rs builds.
If you believe this is docs.rs' fault, open an issue.
Agent memory subsystem -- semantic search over workspace memory files.
Storage: redb (metadata) + hnsw_rs (vector index) under data_dir/memory/.
Embeddings: Configurable via MemorySearchConfig:
- "local" (default): BGE-Small-en-v1.5 via candle-transformers (mmap). Falls back to deterministic pseudo-embedding (FNV hash projection) when model weights are unavailable.
- "openai": OpenAI text-embedding-3-small/large via REST API.
- "ollama": Ollama nomic-embed-text (or custom) via local REST API.
Schema: id(str), scope(str), kind(str), text(str), vector(f32 x DIM). DIM varies by provider: BGE-Small=384, OpenAI-small=1536, etc.