Skip to content

Config reference: graph_storage

  • Enterprise tuning surface

    Defaults + constraints are rendered directly from Pydantic.

  • Env keys when available

    Many fields have an env-style alias (from TriBridConfig.to_flat_dict()).

  • Tooltip-level guidance

    If a matching glossary entry exists, you’ll see deeper tuning notes.

Config reference Config API & workflow Glossary

Total parameters: 14

Group index
  • (root)

(root)

JSON key Env key(s) Type Default Constraints Summary
graph_storage.community_algorithm GRAPH_COMMUNITY_ALGORITHM Literal["louvain", "label_propagation"] "louvain" allowed="louvain", "label_propagation" Community detection algorithm
graph_storage.entity_types — list[str] ["function", "class", "module", "variable", "import"] — Entity types to extract and store in graph
graph_storage.graph_search_top_k GRAPH_SEARCH_TOP_K int 30 ≥ 5, ≤ 100 Number of results from graph traversal
graph_storage.include_communities GRAPH_INCLUDE_COMMUNITIES bool true — Include community detection in graph analysis
graph_storage.max_hops GRAPH_MAX_HOPS int 2 ≥ 1, ≤ 5 Maximum traversal hops for graph search
graph_storage.neo4j_auto_create_databases — bool true — Automatically create per-corpus Neo4j databases when missing (Enterprise).
graph_storage.neo4j_database NEO4J_DATABASE str "neo4j" — Neo4j database name
graph_storage.neo4j_database_mode — Literal["shared", "per_corpus"] "shared" allowed="shared", "per_corpus" Database isolation mode: 'shared' uses a single Neo4j database (Community-compatible), 'per_corpus' uses a separate Neo4j database per corpus (Enterprise multi-database).
graph_storage.neo4j_database_prefix — str "tribrid_" — Prefix for per-corpus Neo4j database names when neo4j_database_mode='per_corpus'.
graph_storage.neo4j_password NEO4J_PASSWORD str "" — Neo4j password (defaults to NEO4J_PASSWORD env var when unset)
graph_storage.neo4j_uri NEO4J_URI str "bolt://localhost:7687" — Neo4j connection URI (bolt:// or neo4j://)
graph_storage.neo4j_user NEO4J_USER str "neo4j" — Neo4j username
graph_storage.neo4j_vector_query_mode NEO4J_VECTOR_QUERY_MODE Literal["auto", "procedure", "search"] "auto" allowed="auto", "procedure", "search" Neo4j chunk-vector query mode. 'auto' prefers runtime-safe defaults and only uses SEARCH where supported.
graph_storage.relationship_types — list[str] ["calls", "imports", "inherits", "contains", "references"] — Relationship types to extract

Details (glossary)

graph_storage.graph_search_top_k (GRAPH_SEARCH_TOP_K) — Graph Search Top-K

Category: general

Number of candidate results to retrieve from Neo4j graph search before fusion. Higher values (40-100) improve recall for graph-based relationships but increase query latency. Lower values (15-30) are faster but may miss relevant connections. Must be >= FINAL_K. Recommended: 30 for balanced performance.

Sweet spot: 30 for production systems. Use 40-50 when graph relationships are critical (e.g., finding code that calls or imports specific functions). Use 15-20 for cost-sensitive scenarios or when graph indexing is sparse.

• Range: 5-100 (typical: 20-50) • Balanced: 30 (recommended) • High recall: 40-50 (relationship-heavy queries) • Cost-sensitive: 15-20 (faster, lower cost) • Effect: Higher = more graph candidates, better recall, higher latency • Symptom too low: Relevant graph connections missed • Symptom too high: Slower queries, diminishing returns

Badges: - Affects latency - Graph relationships

Links: - Neo4j GraphRAG - Graph Traversal - Top-K Retrieval

graph_storage.include_communities (GRAPH_INCLUDE_COMMUNITIES) — Include Communities

Category: general

Enable community-based expansion in graph search. When enabled, the system uses community detection algorithms (e.g., Louvain) to identify clusters of related nodes and expands search to include entire communities. This improves recall for related concepts but may introduce noise. Recommended: enabled for entity mode, optional for chunk mode.

Sweet spot: enabled for entity mode, disabled for chunk mode. Community expansion works best with entity-based graphs where structural clusters are meaningful. For chunk mode, neighbor expansion is usually sufficient.

• Enabled: Community-based expansion, better recall, may introduce noise • Disabled: Direct neighbor expansion only, more focused results • Effect: Controls whether community detection influences traversal • Symptom if disabled: Related concepts in same community may be missed

Badges: - Advanced

Links: - Louvain Algorithm - Community Detection - Community Detection Algorithms

graph_storage.max_hops (GRAPH_MAX_HOPS) — Graph Max Hops

Category: general

Maximum number of graph traversal hops from seed nodes. Each hop expands the search to connected nodes (chunks, entities, relationships). Higher values (3-5) find more distant relationships but increase query latency and may introduce noise. Lower values (1-2) are faster and more focused. Recommended: 2 for balanced performance.

Sweet spot: 2 for production systems. Use 1 for fast, focused traversal (immediate neighbors only). Use 3-4 when you need to find distant relationships or explore deep code structures. Use 5 only for exploratory queries where completeness matters more than speed.

• Range: 1-5 (typical: 1-3) • Focused: 1 (immediate neighbors only) • Balanced: 2 (recommended) • Deep exploration: 3-4 (distant relationships) • Effect: Higher = more relationships explored, better recall, higher latency • Symptom too low: Relevant connections missed • Symptom too high: Slower queries, noise introduced

Badges: - Performance

Links: - Neo4j Graph Traversal - Graph Traversal Depth - Graph Algorithms

graph_storage.neo4j_uri (NEO4J_URI) — Neo4j Connection URI

Category: infrastructure

Connection URI for Neo4j graph database. Used for entity relationships, graph traversal, and community detection in tri-brid search. Format: bolt://host:7687 or neo4j://host:7687. Enables graph-based retrieval for code navigation.