“agent memory”

INVOKE THIS SKILL when your Deep Agent needs memory, persistence, or filesystem access.

Use when adding or reviewing memory search, memory writes, promotion, contradiction, or recall behavior.

Memory is the cornerstone of intelligent agents.

Working, episodic, and semantic memory architectures for autonomous multi-turn agents and long-running sessions.

Scoped CLAUDE.md memory system that reduces context token spend.

Consolidate Codex project/session memory across work sessions.

Operate the Agent Memory Boundary V8 pack with explicit v8 council, ladder, and autonomy boundaries.

Operate the Agent Memory Boundary V8 pack with explicit v8 council, ladder, and autonomy boundaries.

Manage long-term persistence and retrieval of agent context.

Three-tier agent memory model (hot/cold/wiki) for context reduction per spawn

Semantic memory search across all agent memories and conversation history.

Indestructible agent memory — permanently stored, never lost.

Audit the project's persistent agent memories for backfire-risk, stale/superseded state, contradictions, redundancy, and bloat.

Guides memory stitching through indexed retrieval, evidence-gated persistence, and compaction rather than naive transcript accumulation.

Design, audit, and improve agent memory systems across Claude Code's three memory layers (CLAUDE.md, auto-memory, rules/).

Use when a plan finishes or agent memories have piled up.

Implement tiered agent memory: short-term working context, sliding window buffers, episodic vector retrieval, and persistent key-value state.

Decomposes and reorganizes agent memory files into focused, single-purpose components.

Use to extend an existing agent project with memory, app integration, VPC, multi-agent, migration, model, browser, code interpreter, payments, or resource…

AI agent memory with mathematical foundations.

Comprehensive guide for initializing or reorganizing agent memory into a deeply hierarchical file structure.

Institutional memory agent tools.

Build agent memory patterns that prevent the same mistakes — using session replay, error tracking, and a self-correcting knowledge base.

Prepare, review, commit, verify, export, delete, or freeze durable agent memory under explicit consent and read-back control.