ImageFirmAgent engineeringA-06
SheetA-06
Added2026-08
Source1 blueprint taxonomy
MethodVs research & product usage

Seven Kinds of Agent Memory, Checked

A seven-part taxonomy of agent memory has been circulating, and it deserves the compliment this series reserves for its rarest verdict: it is substantially correct. The seven categories map onto how the agent-memory literature and shipping products actually divide the problem. What the poster undersells is the cost side — every kind of memory you add is a new way for your agent to be confidently wrong about stale facts.

01 — The seven, restated and checked

02 — The build order, keptStart small; every layer is a liability

The poster's practical takeaway is genuinely good engineering advice and we keep it verbatim in spirit: start with working memory; add semantic memory when users expect to be remembered; layer episodic, procedural and prospective only when the agent must plan ahead, learn from failure, and adapt. The addition it needs: each layer also adds a failure mode — stale facts served confidently, old lessons applied to new situations, retrieval surfacing the wrong document with authority. Memory without expiry dates and provenance is how agents become confidently wrong. Budget maintenance, not just storage.

03 — The ledgerChecked 2026-08-06

ClaimVerdictThe record
The seven-category taxonomy itselfVERIFIEDAligns with the cognitive-architecture framing used in agent-memory research and products.
Retrieve → reason → act → learn as the cycle memory servesVERIFIEDStandard agent loop, correctly positioned.
Working memory = the context windowVERIFIEDExactly right, including its short-term nature.
RAG classed as a memory typeVERIFIEDFair and useful framing.
Prospective memory as a distinct, practical classNUANCEReal (scheduled goals ship today); the research label is newer than the poster implies.
Incremental build order (start with working memory)VERIFIEDSound engineering; adopted above.
Memory as pure upsideNUANCEThe poster never says it — but never prices staleness, either. Every layer is also a liability.
A-06 · SEVEN KINDS OF AGENT MEMORY, CHECKED · ADDED 2026-08 · ORIGINAL WORK, SOURCES CHECKED 2026-08-06 · VERIFIED SERIES No. 10
PRIMARY SOURCES · agent-memory literature (episodic/semantic/procedural framing) · shipping memory features and scheduled-goal mechanisms (docs.claude.com; harness documentation) · companion sheets A-05, A-07, C-11
BLUEPRINT SOURCE · one third-party seven-type memory poster, on file in the private register — analyzed, not reproduced