agent-memory-systems · diff
git:20260905.f2e5074 to git:20261014.b187a95
10 added, 16 removed. Audit A to A.
---
name: agent-memory-systems
- description: "Memory is the cornerstone of intelligent agents. Without it, every
- interaction starts from zero. This skill covers the architecture of agent
- memory: short-term (context window), long-term (vector stores), and the
- cognitive architectures that organize them."
+ description: "Design tenant-scoped memory, retrieval and retention for agents. Use when planning RAG, conversation history, vector stores or remembering across sessions."
risk: safe
source: vibeship-spawner-skills (Apache 2.0)
date_added: 2026-02-27
---
# Agent Memory Systems
- Memory is the cornerstone of intelligent agents. Without it, every interaction
- starts from zero. This skill covers the architecture of agent memory: short-term
- (context window), long-term (vector stores), and the cognitive architectures
- that organize them.
+ Choose semantic records for facts/preferences, episodic records for events, and procedural records for reusable methods. Keep retrieval quality separate from storage volume and bind persistent access to verified identity.
- Key insight: Memory isn't just storage - it's retrieval. A million stored facts
- mean nothing if you can't find the right one. Chunking, embedding, and retrieval
- strategies determine whether your agent remembers or forgets.
+ ## Procedure
- The field is fragmented with inconsistent terminology. We use the CoALA cognitive
- architecture framework: semantic memory (facts), episodic memory (experiences),
- and procedural memory (how-to knowledge).
+ 1. Establish the actual task, authenticated tenant/user, permission to retain content, retention limits and existing store/model versions.
+ 2. Read the guide's tenant-scoped contract and relevant storage/chunking sections. Select an adapter supported by the installed SDK; the guide's historical sketches are not deployable integrations.
+ 3. Specify record IDs, source evidence, timestamps, namespace, retrieval limits and conflict/version policy. Preserve originals and propose consolidation without deletion.
+ 4. Test cross-identity rejection, stale/conflicting records, bounded retrieval and failed writes with isolated fixtures. Benchmark retrieval against held-out questions before expanding storage.
+ 5. Deliver the architecture and evidence-backed change plan, listing unresolved adapters and hosted cost/data destinations. Apply only the persistence or provider operations actually authorized.
## Detailed Guide
Read [the detailed guide](references/detailed-guide.md) before executing this skill. It retains the complete procedure and reference material. Treat its safety, prerequisites, and validation requirements as mandatory. For focused work, load the relevant sections; for end-to-end work, read the guide completely.
## When to Use
- User mentions or implies: agent memory
- User mentions or implies: long-term memory
- User mentions or implies: memory systems
- User mentions or implies: remember across sessions
- User mentions or implies: memory retrieval
- User mentions or implies: episodic memory
- User mentions or implies: semantic memory
- User mentions or implies: vector store
- User mentions or implies: rag
- User mentions or implies: langmem
- User mentions or implies: memgpt
- User mentions or implies: conversation history
## Example
- **User request:**
+ **Request:** “Our support agent forgets preferences between sessions. Design memory for two customer tenants without sharing their history.”
- > Use @agent-memory-systems for this task: Memory is the cornerstone of intelligent agents.
+ **Output:** Load the tenant-scoped contract and background-extraction sections. Propose a semantic namespace per verified tenant/user, records with source IDs and revisions, and bounded retrieval. Retain conflicting preferences until the owner resolves them; include tests that tenant B cannot retrieve or consolidate tenant A records. Identify the existing persistent adapter and consent/retention evidence before any write.
## Limitations
- Use this skill only when the task clearly matches the scope described above.
- Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
- Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.