robotmem
robotmem is a persistent memory system for robotic/agent experiences. It records episodic robot experience (parameters/strategies/trajectories/outcomes), stores perception data, and retrieves relevant past memories using hybrid search (BM25 + vector) with structured JSON filtering and optional spatial nearest-neighbor sorting. It can also run as an MCP server (7 tools) or be used directly via Python imports, with an embedded local SQLite database.
Score Breakdown
⚙ Agent Friendliness
🔒 Security
Appears to be local/offline-first using embedded SQLite, which reduces network attack surface, but no auth model, TLS requirements, or secret-handling/logging guarantees are documented. It includes an MCP server and optional web management UI (flask), but security configuration (auth, CSRF, TLS, CORS, rate limiting) is not documented in the provided materials.
⚡ Reliability
Best When
You have robotics/agent experiments where you can run a local persistent SQLite-backed memory store and want structured + spatially informed retrieval to guide future decisions.
Avoid When
You need a hosted service with enterprise-grade security, centralized access control, and documented SLAs, or you cannot store data locally.
Use Cases
- • Improving robot policies via experience replay (record → recall across episodes)
- • Retrieving only relevant prior episodes using structured constraints (e.g., task success, parameter thresholds)
- • Spatial/nearest-neighbor retrieval for scenarios with similar physical positions
- • Offline-first local long-term memory for robotics experiments (CPU-only, SQLite-backed)
- • Agent workflows that benefit from hybrid retrieval (text + embeddings) fused with reranking
Not For
- • Production multi-tenant SaaS scenarios requiring managed hosting, auth, and auditability
- • Use cases that require network-based low-latency shared memory at scale across many users
- • Environments that require strong enterprise compliance controls (SOC2/HIPAA/etc.) out of the box
Interface
Authentication
No authentication mechanisms are described in the provided README/manifest. The system appears to be local/offline-first (embedded SQLite, CPU-only) and thus likely used without external auth unless the MCP/web UI adds it (not documented here).
Pricing
Apache-2.0 library; no pricing details provided.
Agent Metadata
Known Gotchas
- ⚠ No documented rate limiting behavior (local usage likely, but MCP/UI could differ).
- ⚠ Structured filtering expects specific JSON field paths and operators (e.g., $lt); incorrect schema/paths may lead to empty/incorrect recall without clear guidance.
- ⚠ Spatial sorting depends on consistent coordinate field naming and the expected 'spatial.*' structure; mismatches may reduce retrieval quality.
Alternatives
Full Evaluation Report
Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for robotmem.
AI-powered analysis · PDF + markdown · Delivered within 30 minutes
Package Brief
Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.
Delivered within 10 minutes
Score Monitoring
Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.
Continuous monitoring
Scores are editorial opinions as of 2026-03-30.