From agentdb-memory
Store a memory in AgentDB — an episode (task + outcome + critique), a pattern, or a skill. Use when the user says "remember this", "save this for later", "add to memory", or when the agent has just succeeded/failed at a task and the lesson is worth keeping.
npx claudepluginhub ruvnet/agentdb --plugin agentdb-memoryThis skill uses the workspace's default tool permissions.
Persist a memory into AgentDB so a future session can recall it. Three flavors — pick the right one for what's being remembered:
Retrieve relevant memories for the current task from AgentDB. Use at the start of a task to load prior knowledge, when stuck to surface what worked before, or when the user asks "what do we know about X" / "have we done this before?"
Saves insights, decisions, or learnings to agentmemory long-term storage via curl POST API. Extracts concepts and files; activates on 'remember this' or 'save this'.
Manages local memory for AI agents with compression detection, auto-snapshots, semantic search across episodic, semantic, and procedural tiers. Use to prevent context loss, organize memories, and track usage.
Share bugs, ideas, or general feedback.
Persist a memory into AgentDB so a future session can recall it. Three flavors — pick the right one for what's being remembered:
| User phrasing | Use |
|---|---|
| "remember that X works for Y" | Pattern — agentdb_pattern_store |
| "save what happened when I tried X" | Episode (Reflexion) — agentdb_reflexion_store |
| "this is a reusable approach for problem X" | Skill — agentdb_skill_create |
agentdb_pattern_store(
content: "JWT refresh token rotation pattern: ...",
metadata: { topic, project, success: true }
)
Use for facts, conventions, anti-patterns, or anything that should resurface as a hint when a similar query comes up later.
agentdb_reflexion_store(
sessionId: <session>,
task: <what we were trying to do>,
input: <what we tried>,
output: <what happened>,
critique: <what we'd do differently>,
reward: 0..1,
success: true|false
)
Use right after a task completes — success OR failure. Failed episodes feed getCritiqueSummary; successful ones feed getSuccessStrategies.
agentdb_skill_create(
name: <short verb-noun>,
description: <what it does>,
precondition: <when to use>,
action: <how to do it>,
outcome: <what success looks like>
)
Use when a pattern has been validated 3+ times and you want it elevated to a first-class reusable skill (queryable by intent embedding).