Agent Memory Systems

Memory systems are where agent architectures get real. This skill breaks down the three-layer approach: semantic memory for facts and preferences, episodic for timestamped experiences, and procedural for learned patterns. The tooling coverage is practical, comparing Pinecone's enterprise scale against ChromaDB for prototyping, with honest latency numbers and cost tradeoffs. The core insight holds up: retrieval quality matters more than storage quantity, which is why the chunking and embedding strategy sections matter. Built on the CoALA cognitive framework, so the terminology is consistent even if the broader field isn't. If you're building agents that need to remember across sessions, this covers the architecture decisions you'll actually face.

skills.sh installs: 1.7K
GitHub stars: 43.1K
last updated: Jul 14, 2026

View on GitHub

Install

$ npx -y skills add sickn33/antigravity-awesome-skills --skill agent-memory-systems --agent claude-code

Installs into .claude/skills of the current project.

Files

Agent Memory Systems

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.

Procedure

Establish the actual task, authenticated tenant/user, permission to retain content, retention limits and existing store/model versions.
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.
Specify record IDs, source evidence, timestamps, namespace, retrieval limits and conflict/version policy. Preserve originals and propose consolidation without deletion.
Test cross-identity rejection, stale/conflicting records, bounded retrieval and failed writes with isolated fixtures. Benchmark retrieval against held-out questions before expanding storage.
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.