Stats
Actions
Tags
From agentdb-causal
Record a causal link between two memories — A causes B, A supersedes B, A depends on B.
How this command is triggered — by the user, by Claude, or both
Slash command
/agentdb-causal:linkThe summary Claude sees in its command listing — used to decide when to auto-load this command
Use the `agentdb-causal-link` skill to add a typed edge between two memories. If the user provides explicit `(from, to, relation)` after `/link`, store that. Otherwise infer from the most recent context — the last failed test that the most recent fix targeted, the ADR that supersedes the previous one, etc. — and confirm before storing. After storing, briefly tell the user what edge was added and offer to record uplift / confidence if they have evidence.
Use the agentdb-causal-link skill to add a typed edge between two memories.
If the user provides explicit (from, to, relation) after /link, store that. Otherwise infer from the most recent context — the last failed test that the most recent fix targeted, the ADR that supersedes the previous one, etc. — and confirm before storing.
After storing, briefly tell the user what edge was added and offer to record uplift / confidence if they have evidence.
npx claudepluginhub ruvnet/agentdb --plugin agentdb-causal/memoryManages long-term memories – search, store, forget, link, and review proposals. Supports recall, learn, forget, relate, and propose subcommands.
/rememberStores decisions, patterns, outcomes, and context in a knowledge graph for future reference. Supports flags for success/failure tracking, categories, agent scoping, and global best practices.
/graphDisplays related memories as an indented knowledge-graph tree up to 3 hops from a given ID or prefix, showing relationship types and directions (e.g., → supersedes, ← derived_from).
/memory-saveSaves current conversation context as atomic memory in Forgetful. Queries existing memories, drafts curated entry with title/content/context/keywords/tags/importance, proposes obsoleting/linking actions, and executes on confirmation.
/captureCaptures a new memory to the mnemonic filesystem with required namespace and title, optional type (semantic/episodic/procedural), tags, scope (project/org), confidence, citations. Validates against ontology.