General-purpose memory layers
Hosted or open-source memory APIs (Mem0, Letta, Zep, Honcho and others). Services and libraries that store memories for an agent, usually as extracted facts, memory blocks or a knowledge graph, and retrieve them by query.
Verdict
Solid for agents you build yourself; for off-the-shelf coding agents, check how they capture sessions and what they throw away.
Best for
Teams building their own agent or product who want a memory API to call, and can wire capture in themselves.
Poor fit
Teams who want Claude Code, Codex and Cursor sessions captured without writing integration code, or who need the original wording of a decision.
Scores
| Criterion | Score | Why |
|---|---|---|
| Survival across sessions | Strong | Stored outside the session by design. |
| Cross-tool reach | Partial | Reachable from anything that calls the API or MCP server, but capture from each coding agent varies; check the integrations list. |
| Multi-agent sharing | Partial | Agents can share one store, but awareness of what others are doing right now is not standard. |
| Retention of reasons | Partial | Depends on what is stored; extracted facts and summaries tend to lose the why. |
| Scale | Partial | Retrieval-based, but published BEAM scores vary widely between products and fall at 10M tokens (see /benchmarks). |
| Openness | Partial | Several are open source; hosted tiers and features differ. Check each licence. |
Strengths
- Context lives outside the session, so it survives restarts and new machines.
- Several have open-source cores you can read and run (check each repository's licence).
- Built to scale past a context window by retrieving only what is relevant.
Limits
- Many store what an extraction step judged important, not the conversation; the reason for a decision can be the detail that is dropped.
- Built for agents in general, so capture from a given coding agent may need code or a plugin you set up.
- Live coordination between agents (who is changing what right now) is usually not what they are built for.
How it fails
- An extraction step decides at write time what will matter later, and gets it wrong for the question you ask next week.
- Contradictory memories from different sessions sit side by side with nothing to say which is current.
Cost
Varies by product and tier; each publishes its own pricing. Check mem0.ai, letta.com, getzep.com and plasticlabs.ai before you compare.