Read-only inspection of the knowledge graph — no side effects.
Operation and target: $ARGUMENTS
consolidate(action: "query", operation: "list_nodes", node_type: "semantic", min_confidence: 0.5, limit: 20)Browse what's stored. Filter by node_type (semantic, procedural, episodic, temporal, outcome, goal), min_confidence, max_confidence, limit.
consolidate(action: "query", operation: "get_node", node_id: "<id>")Returns: content, confidence, source, metadata, node_type, timescale (fast/medium/slow/glacial), access_count, created_at, updated_at.
consolidate(action: "query", operation: "get_edges", node_id: "<id>")Returns both outgoing and incoming edges with types and weights. Useful for tracing causal chains, finding contradictions, and understanding provenance.
consolidate(action: "query", operation: "similarity_search", query: "<text>", limit: 5)Embedding-based search. Returns nodes with similarity scores.
Operation aliases: "get", "edges", "list" also work as shorthand.
consolidate(action: "traverse", start_node_id: "<id>", max_depth: 3, relationship_types: "causal,supports")BFS walk from a starting node. Returns visited nodes, edges, depth map, and traversal metadata. Optionally filter by relationship types (comma-separated).
consolidate(action: "stats")Returns node/edge counts, type distributions, confidence statistics (mean/min/max/std_dev), and orphan count (nodes with no edges). No parameters required.
act(action: "manage_edge", operation: "list_all")
act(action: "manage_edge", operation: "list_for_node", node_id: "<id>")
act(action: "manage_edge", operation: "update", edge_id: "<id>", weight: 0.9, co_activation_count: 5)
act(action: "manage_edge", operation: "delete", edge_id: "<id>")Full edge lifecycle management: list, update weight/co_activation_count/decay_rate, or delete.
act(action: "delete_node", node_id: "<id>")Remove a node and its connected edges from the graph.
| URI | What It Returns |
|---|---|
graphonomous://runtime/health | Runtime health, service status, lightweight counts |
graphonomous://goals/snapshot | Goal totals, status breakdown, serialized goals |
graphonomous://graph/node/{id} | Individual node details + connected edges |
graphonomous://graph/recent | Recently added/accessed nodes, sorted by recency |
graphonomous://consolidation/log | Consolidator state + orchestrator plasticity metrics |
retrieve(action: "context") vs consolidate(action: "query", operation: "similarity_search")retrieve(action: "context") | consolidate(action: "query", operation: "similarity_search") | |
|---|---|---|
| Neighborhood expansion | Yes (hops) | No |
| Topology annotations | Yes (routing, kappa) | No |
| Causal context | Yes (for outcome learning) | No |
| Side effects | Updates access counts | None |
| Use when | Acting on results (need causal tracking) | Checking duplicates, auditing, browsing |
Rule: Use retrieve(action: "context") when you plan to act on results. Use consolidate(action: "query") for read-only inspection.
consolidate(action: "query", operation: "similarity_search", query: "the fact I want to store", limit: 3)similarity > 0.90 — duplicate — update the existing node instead
similarity 0.70-0.90 — related — store new node + add related edge
similarity < 0.70 — novel — safe to store as new node
consolidate(action: "query", operation: "list_nodes", min_confidence: 0.0, max_confidence: 0.3)Find speculative or likely-wrong nodes needing reinforcement or pruning.
consolidate(action: "query", operation: "get_node", node_id: "<id>")
consolidate(action: "query", operation: "get_edges", node_id: "<id>")Follow causal and supports edges to reconstruct reasoning. Check for contradicts edges that indicate unresolved conflicts.
Using retrieve(action: "context") for browsing — it has side effects (updates access counts)
Forgetting to check for duplicates before storing new nodes
Running traversals with very high max_depth on large graphs — start with 2-3