Expand description
§Terraphim Knowledge Graph Agents
Specialized GenAgent implementations that leverage knowledge graph capabilities for intelligent task planning, execution, and coordination.
This crate provides concrete implementations of the GenAgent trait that integrate deeply with Terraphim’s knowledge graph infrastructure to provide:
- Planning Agents: Intelligent task decomposition and execution planning
- Worker Agents: Domain-specialized task execution with knowledge graph context
- Coordination Agents: Multi-agent workflow coordination and supervision
§Core Features
- Knowledge Graph Integration: Deep integration with automata and role graphs
- Domain Specialization: Agents specialized for specific knowledge domains
- Task Compatibility: Intelligent task-agent matching using connectivity analysis
- Context-Aware Execution: Task execution guided by knowledge graph context
- Coordination Capabilities: Multi-agent workflow orchestration
Re-exports§
pub use error::*;
Modules§
- capability
- Capability-based routing types for unified LLM and Agent providers.
- error
- Error types for knowledge graph agents
- llm_
usage - LLM usage tracking types for cost monitoring across providers.
- mcp_
tool - MCP Tool types for indexing and discovery.
- persona
- Persona definition types for agent personas with SFIA skill framework support.
- procedure
- Procedure capture types for the learning system.
- review
- Review finding types for multi-agent code review.
- score
- Scoring algorithms and query types for document relevance ranking. Scoring algorithms and query types for document relevance ranking.
Structs§
- Agent
Communication - Communication between agents in a multi-agent context
- Agent
Info - Information about an AI agent in a multi-agent context
- Agent
Metadata - Comprehensive agent metadata
- Agent
Pid - Unique identifier for agents in the supervision system
- Captured
Procedure - A captured procedure with ordered steps and execution history.
- Characteristic
Def - Definition of a core personality characteristic.
- Chat
Message - A single message in a conversation, including metadata for cost tracking.
- Concept
- A concept is a higher-level, normalized term.
- Context
History - Context history that tracks what context has been used across conversations
- Context
History Entry - Entry in context usage history
- Context
Item - Context item that can be added to LLM conversations
- Conversation
- Conversation
Id - Unique identifier for conversations
- Conversation
Summary - Summary of a conversation for listing purposes
- Coverage
Signal - Coverage governance signal
- Document
- The central document type representing indexed and searchable content.
- Edge
- A directed relationship between two nodes in the knowledge graph.
- Extracted
Entity - Extracted entity from text
- Extracted
Relationship - Extracted relationship from text
- Grounding
Metadata - Grounding metadata for normalized terms (Dynamic Ontology)
- Index
- An index is a hashmap of documents
- Indexed
Document - Reference to external storage of documents
- KGIndex
Info - Knowledge Graph index information
- KGTerm
Definition - Knowledge Graph term definition with comprehensive metadata
- LlmResult
- The text content produced by an LLM call together with optional usage metadata.
- LlmUsage
- Token usage and cost data for a single LLM call.
- Markdown
Directives - Parsed directives extracted from the YAML front matter of a markdown KG entry.
- McpTool
Entry - Represents an indexed MCP tool from configured servers.
- Message
Id - Unique identifier for messages within conversations
- Model
Pricing - Per-model pricing configuration for cost estimation.
- Multi
Agent Context - Multi-agent context for coordinating between different AI agents
- Node
- A
Noderepresents single concept and its connections to other concepts. - Normalized
Term - A normalized term is a higher-level term that has been normalized
- Normalized
Term Value - The value of a normalized term
- Ontology
Anti Pattern - Anti-pattern definition for detection
- Ontology
Entity Type - Entity type definition in an ontology schema
- Ontology
Relationship Type - Relationship type definition in an ontology schema
- Ontology
Schema - Schema-first ontology definition
- Pattern
Match - Result of pattern matching with priority scoring
- Persona
Definition - A complete persona definition for an AI agent.
- Priority
- Priority level for routing rules and decisions Higher numeric values indicate higher priority A clamped priority value in the range 0–100 (higher = more urgent).
- Procedure
Confidence - Confidence metrics for a procedure based on execution history.
- Procedure
Step - A single step in a captured procedure.
- Process
Id - Process ID for spawned agents
- Provider
- A provider that can fulfill capabilities
- Quality
Score - Quality scores for Knowledge/Learning/Synthesis (K/L/S) dimensions.
- Review
Agent Output - Output schema for a single review agent’s results.
- Review
Finding - A single structured finding from a review agent.
- Role
Name - A role name with case-insensitive lookup support.
- Route
Directive - Routing directive specifying which LLM provider and model to use.
- Routing
Decision - Routing decision with priority information
- Routing
Rule - A routing rule with pattern matching and priority
- Schema
Signal - Schema signal extracted from text
- Search
Query - A search query for finding documents in the knowledge graph.
- Sfia
Skill Def - SFIA skill definition.
- Supervisor
Id - Unique identifier for supervisors
- Thesaurus
- A thesaurus is a dictionary with synonyms which map to upper-level concepts.
Enums§
- Capability
- A capability that a provider can fulfill
- Context
Type - Types of context that can be added to conversations
- Context
Usage Type - How a context item was used
- Cost
Level - Cost level for routing decisions
- Document
Type - Classifies a document by its role in the knowledge graph.
- Finding
Category - Category of a review finding (maps to review groups).
- Finding
Severity - Severity level for a review finding.
- GenAgent
Error - Knowledge
Graph Input Type - Defines all supported inputs for the knowledge graph.
- Latency
- Latency expectation for routing decisions
- Layer
- Layered output levels for search results.
- Logical
Operator - Logical operators for combining multiple search terms
- Normalization
Method - Normalization method used for grounding
- Persona
Load Error - Errors that can occur when loading a persona definition.
- Provider
Type - Type of provider (LLM or Agent)
- Relevance
Function - Defines the relevance function (scorer) to be used for ranking search
results for the
Role. - RotStatus
- Health status of conversation context based on token budget utilization.
- Routing
Scenario - Routing scenario types
Functions§
- deduplicate_
findings - Deduplicate findings by keeping the most severe finding per (file, line, category) key.
- extract_
first_ paragraph - Extract the first paragraph from document body text.
Type Aliases§
- GenAgent
Result - KgAgent
Result - Result type for knowledge graph agent operations