// Execute Cypher queries against AgentDB's graph backend. Use when the user wants to write a custom traversal that the standard tools don't cover, or when explaining graph state.
| name | agentdb-cypher |
| description | Execute Cypher queries against AgentDB's graph backend. Use when the user wants to write a custom traversal that the standard tools don't cover, or when explaining graph state. |
Run arbitrary Cypher against the AgentDB graph backend. The native binding accepts standard openCypher; AgentDB also adds a few extensions (uplift, confidence properties on edges).
agentdb_causal_* tools don't coveragentdb_causal_query(
cypher: <query string>,
params?: { ... }
)
Returns: { rows, stats: { execMs, rowsMatched, rowsReturned } }
-- Top 10 most-cited skills
MATCH (s:Skill)<-[r:references]-(:Episode)
RETURN s.name, count(r) AS cites
ORDER BY cites DESC LIMIT 10
-- Find hub nodes (high in-degree)
MATCH (n)<-[r]-()
RETURN n.id, n.type, count(r) AS inDegree
ORDER BY inDegree DESC LIMIT 20
-- Walk supersedes chain backwards from current ADR
MATCH p=(current:ADR {id: 'ADR-046'})-[:supersedes*]->(ancestor:ADR)
RETURN [n IN nodes(p) | n.title]
-- Cycle detection (worth a warning)
MATCH p=(n)-[:caused*1..5]->(n)
RETURN p LIMIT 5
The agentdb_causal_query tool accepts arbitrary Cypher, including writes (CREATE, DELETE, DETACH DELETE). For destructive operations, prefer the typed delete tools (agentdb_causal_node_delete, agentdb_causal_edge_delete) — they enforce cascade semantics and write to the attestation log.
params to avoid Cypher injection.MATCH (n) DETACH DELETE n (deletes everything). Always scope the match.WHERE id IN [...] works — the binding's per-call overhead adds up.Walk the causal graph in AgentDB to explain why two memories are connected, or trace a root cause. Use when the user asks "why did X happen", "what led to Y", or after an incident.
Record a causal relationship between two memories in AgentDB — "X caused Y", "A supersedes B", "patch-foo depends-on patch-bar". Use when the user is documenting cause/effect, dependencies, supersessions, or after-action analysis.
Store memories in tier-aware hierarchical memory — working / short-term / long-term — and recall with tier filters. Use for working-set context that should fade, vs facts that should persist, vs patterns that should be searchable forever.
Initialize an AgentDB Cognitive Container (.rvf file) in the current project. Sets up storage, embedder config, and the agentdb MCP server. Use when the user is starting a new project that needs vector memory, or asks to "set up agentdb" / "init agentdb".
Show AgentDB health — pattern count, embedder status, cache hit rate, learning gain since init. Use when the user asks "is agentdb working?", "how many memories?", "show agentdb stats", or after long-running sessions to confirm state.
Search and manage hyperedges — n-ary relationships between memories. Use for swarm membership, multi-cause incidents, or any "this involves all of (A, B, C, D)" relationship that doesn't fit a binary edge.