Long-term episodic agent memory

Scores as of 2026-09-14 (43 days old; Depot marks scores stale after 30 days). The pick is still Depot's current editorial pick; it has not been re-reviewed since. Prices below carry their own dates.

Depot Pick

Zep · 7.48/10 for this job (curated estimate, high confidence)

Distinct from `hosted-agent-memory` / `temporal-graph-memory` nuance: episodic timelines. Confidence: medium.

Runner-up

Letta Cloud · 7.37/10

Scores higher than the pick on: Price / performance, Developer & agent DX.

When not to use the pick

Mem0 simplicity, Letta stateful runtime, or HIPAA memory (Supermemory) is the constraint.

Lock-in / migration

The pick has lower stored lock-in risk than runner-up Letta Cloud (5.2 vs 5.5). Pick exit signals: open source (provider tag `oss`). Self-host fallback listed for this job: Graphiti.

Sourced price (no comparable unit)

Pick: no sourced price in this job's unit (no_sourced_price).

No cheapest SKU (no_comparable_unit_for_job). No price was invented.

Free option

• Graphiti (oss) — source, as of 2026-10-25

Failure modes to watch

• Context overflow: If the input exceeds the window, say so and retrieve a narrower slice. Do not answer from a truncated head as if it were the whole source.
• Stale retrieval: Store the source URL and the time it was fetched. If a newer canonical page exists, or two stored sources disagree, do not answer from the older one.
• Confirmation skip: Gate those tool names on an approval id recorded before the call. A missing approval is a failure even if the downstream tool returns success.
• Identity confusion: Bind every tool call to a user or tenant id. Refuse the call if that id changed mid-task. For a public bot, the operator owns the words the bot shows.
• Instruction drift: Restate the stop condition in the trace. End the run when it is met. Review long traces for steps that do not serve the original task.
• Prompt injection: Compare each tool call to the original user task. If the call only makes sense given text that arrived from a fetched page, file, or message, stop and ask. Record which source the instruction came from. This catalog does not store payloads or attack steps.

Reported outcomes (submitted reports only)

This job: loading… JSON · reports never change scores · seed/test records excluded

Depot Pick (Zep) on this job: loading… JSON · reports never change scores · seed/test records excluded

Cite: Agent Provider Depot, Depot Pick for episodic-long-term-memory: zep, scores as of 2026-09-14. Agents: GET /api/decide?jtbd=episodic-long-term-memory