How do I install the Agentdb Memory Patterns skill?

Run npx skillmds@latest add dreamlab-ai/agentdb-memory-patterns in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Agentdb Memory Patterns safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Agentdb Memory Patterns?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Agentdb Memory Patterns free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Agentdb Memory Patterns?

dreamlab-ai (@dreamlab-ai) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Agentdb Memory Patterns by dreamlab-ai

Name: Agentdb Memory Patterns
Author: dreamlab-ai

Agentdb Memory Patterns

Implement persistent memory patterns for AI agents using AgentDB — session memory, long-term storage, pattern learning, hierarchical memory, consolidation, ReasoningBank migration. Use when building stateful agents, chat systems, or intelligent assistants. NOT for distributed multi-DB/QUIC sync or RL algorithms (use agentdb-advanced), vector-search perf tuning (use agentdb-vector-search), or non-persistent session state.

dreamlab-ai 484d7c1 4 files · 20.5 KB Updated Sep 17, 2026

File contents

AgentDB Memory Patterns

AgentDB is a standalone RuvNet package, not this deployment's memory path; durable agent/project memory here is RuVector-postgres via mcp__claude-flow__memory_* (bge-small-en-v1.5, 384-dim) — see KHIVE-LEARNINGS.md §3 for the detailed comparison.

For AgentDB architecture and performance overview, see AgentDB Overview.

What This Skill Covers

Memory management patterns for AI agents: session memory, long-term storage, pattern learning, hierarchical memory organisation, memory consolidation, and ReasoningBank migration. For CLI setup, API initialisation, and common operations, see the AgentDB Overview.

Additional Prerequisites: Understanding of agent architectures.

When Not To Use

For distributed multi-database setups or QUIC sync -- use the agentdb-advanced skill instead
For reinforcement learning algorithms -- use the agentdb-advanced skill (RL Plugins section) instead
For pure vector search performance tuning (quantisation, HNSW) -- use the agentdb-vector-search skill instead
For non-persistent, session-only state that does not need vector search -- standard in-memory data structures suffice
For unified development and quality engineering workflows -- use the build-with-quality skill which includes memory management

Pattern Reference

Full code examples live in references/pattern-examples.md:

Session Memory — per-session message store with time-ordered history.
Long-Term Memory — categorised facts with confidence and source.
Pattern Learning — record successful trigger→response pairs and match them against live context.
Hierarchical Memory — organise across immediate / short-term / long-term / semantic tiers.
Memory Consolidation — periodic importance-based pruning to a size/score threshold.
ReasoningBank Integration — migrate a legacy .swarm/memory.db, train the learning model, and retrieve with reasoning.

For learning plugins and RL algorithms, see the agentdb-advanced skill (RL Plugins section). For performance tuning and quantization, see the agentdb-vector-search skill.

KHIVE-Informed Enhancements (v2.0 Roadmap)

Proposed enhancements from KHIVE v2 comparison — orient cold-start dashboard, importance-weighted storage, entity-relationship graph, task dependency DAG, tag retrieval path, and URI/URN addressing. Full proposal with schemas and code in references/khive-v2-roadmap.md; underlying analysis in KHIVE-LEARNINGS.md.

Troubleshooting

Issue: Memory growing too large

# Check database size
npx agentdb@latest stats ./agents.db

# Enable quantization
# Use 'binary' (32x smaller) or 'scalar' (4x smaller)

Issue: Slow search performance

# Enable HNSW indexing and caching
# Results: <100µs search time

Issue: Migration from legacy ReasoningBank

# Automatic migration with validation
npx agentdb@latest migrate --source .swarm/memory.db

Learn More

For performance benchmarks, see AgentDB Overview. For general links, see AgentDB Overview.

---name: agentdb-memory-patternsdescription: Implement persistent memory patterns for AI agents using AgentDB — session memory, long-term storage, pattern learning, hierarchical memory, consolidation, ReasoningBank migration. Use when building stateful agents, chat systems, or intelligent assistants. NOT for distributed multi-DB/QUIC sync or RL algorithms (use agentdb-advanced), vector-search perf tuning (use agentdb-vector-search), or non-persistent session state.---# AgentDB Memory PatternsAgentDB is a standalone RuvNet package, not this deployment's memory path; durable agent/project memory here is RuVector-postgres via `mcp__claude-flow__memory_*` (bge-small-en-v1.5, 384-dim) — see [KHIVE-LEARNINGS.md §3](KHIVE-LEARNINGS.md) for the detailed comparison.For AgentDB architecture and performance overview, see [AgentDB Overview](../agentdb-advanced/docs/agentdb-overview.md).## What This Skill CoversMemory management patterns for AI agents: session memory, long-term storage, pattern learning, hierarchical memory organisation, memory consolidation, and ReasoningBank migration. For CLI setup, API initialisation, and common operations, see the [AgentDB Overview](../agentdb-advanced/docs/agentdb-overview.md).**Additional Prerequisites**: Understanding of agent architectures.## When Not To Use- For distributed multi-database setups or QUIC sync -- use the agentdb-advanced skill instead- For reinforcement learning algorithms -- use the **agentdb-advanced** skill (RL Plugins section) instead- For pure vector search performance tuning (quantisation, HNSW) -- use the agentdb-vector-search skill instead- For non-persistent, session-only state that does not need vector search -- standard in-memory data structures suffice- For unified development and quality engineering workflows -- use the build-with-quality skill which includes memory management## Pattern ReferenceFull code examples live in [references/pattern-examples.md](references/pattern-examples.md):- **Session Memory** — per-session message store with time-ordered history.- **Long-Term Memory** — categorised facts with confidence and source.- **Pattern Learning** — record successful trigger→response pairs and match them against live context.- **Hierarchical Memory** — organise across immediate / short-term / long-term / semantic tiers.- **Memory Consolidation** — periodic importance-based pruning to a size/score threshold.- **ReasoningBank Integration** — migrate a legacy `.swarm/memory.db`, train the learning model, and retrieve with reasoning.For learning plugins and RL algorithms, see the [agentdb-advanced](../agentdb-advanced/SKILL.md) skill (RL Plugins section). For performance tuning and quantization, see the [agentdb-vector-search](../agentdb-vector-search/SKILL.md) skill.## KHIVE-Informed Enhancements (v2.0 Roadmap)Proposed enhancements from KHIVE v2 comparison — orient cold-start dashboard, importance-weighted storage, entity-relationship graph, task dependency DAG, tag retrieval path, and URI/URN addressing. Full proposal with schemas and code in [references/khive-v2-roadmap.md](references/khive-v2-roadmap.md); underlying analysis in [KHIVE-LEARNINGS.md](KHIVE-LEARNINGS.md).## Troubleshooting### Issue: Memory growing too large```bash# Check database sizenpx agentdb@latest stats ./agents.db# Enable quantization# Use 'binary' (32x smaller) or 'scalar' (4x smaller)```### Issue: Slow search performance```bash# Enable HNSW indexing and caching# Results: <100µs search time```### Issue: Migration from legacy ReasoningBank```bash# Automatic migration with validationnpx agentdb@latest migrate --source .swarm/memory.db```## Learn MoreFor performance benchmarks, see [AgentDB Overview](../agentdb-advanced/docs/agentdb-overview.md#performance-claims). For general links, see [AgentDB Overview](../agentdb-advanced/docs/agentdb-overview.md#links).

Frequently asked questions