Skip to main content
Public BetaWe're in Public Beta. Things may break. Please report issues via the Support tab.

Knowledge Graph

Interactive graph visualization of all 9 entity types with 12 relationship types - auto-generated by AI Builder or manually created

What is the Knowledge Graph?

The Knowledge Graph allows you to create relationships between different entities in ACE, building a network of connected information that mirrors how ideas actually relate in your projects.

Connect Entities

Link memories, decisions, issues, and work logs together to show how they relate

Interactive Visualization

Force-directed graph layout with drag-to-position, zoom, pan, and right-click interactions

Focus Mode

Right-click any node to enter focus mode - filters graph to show only that node and its neighbors

Traverse & Track Context

Explore connections up to 4 levels deep and understand how entities relate across your project

Auto-Generated by AI Builder
Create complete knowledge graphs in seconds with AI-powered plan generation

The AI-Assisted Plan Builder automatically creates comprehensive knowledge graphs when you generate project plans. Simply describe your project, and the AI creates all entities AND their relationships in one atomic operation.

Automatic Relationships

AI creates semantic relationships between issues, decisions, architecture, and tests based on project context

Ready to Visualize

Immediately visualize your entire project structure in the interactive knowledge graph dashboard

9 Entity Types in Knowledge Graph
All ACE entity types can be connected with semantic relationships

Issues

Features to build, bugs to fix, tasks to complete

Decisions

Architectural choices with rationale and alternatives

Work Logs

Daily activities, progress tracking, completed tasks

Architecture

System components, technology stack, data flow

Best Practices

Guidelines and recommended patterns for your team

Patterns

Reusable design patterns and code templates

Tags

Categorization labels for organizing all entities

Relationships

The connections themselves - meta-relationships

Generic Memories

Unstructured knowledge and context not fitting other types

Relationship Types
12 semantic relationship types to express how entities connect
implements

Entity implements a pattern or best practice

references

Entity references another for context

supersedes

Entity replaces an older version

relates_to

General relationship between entities

caused_by

Entity was caused by another (e.g., issue caused by decision)

resolves

Entity resolves a problem or issue

documented_in

Entity is documented in another place

depends_on

Entity has a dependency on another

blocks

Entity blocks progress on another

duplicates

Entity is a duplicate of another

validates

Entity validates another (e.g., test validates architecture)

reproduces

Entity reproduces an issue or scenario

Using the Knowledge Graph Dashboard

1. Access the Dashboard

Navigate to the Knowledge Graph page from your ACE dashboard navigation.

Dashboard → Knowledge Graph

2. Select an Entity

  1. Choose your namespace from the dropdown
  2. Select an entity type (Issues, Decisions, Work Logs, Architecture, Best Practices, Patterns, Tags, Relationships, or Generic Memories)
  3. Search for a specific entity using the search box
  4. Click on an entity from the search results

3. Visualize Relationships

The graph will automatically load and display:

  • Nodes: Colored circles representing entities with unique colors for each type (amber=issues/decisions, green=work logs, blue=architecture, purple=testing, emerald=best practices, pink=patterns, slate=tags, red=relationships, gray=generic memories)
  • Node labels: Descriptive text showing entity name or summary (work logs show "date: description" format)
  • Edges: Arrows showing relationships with type labels (implements, resolves, depends_on, etc.)
  • Depth: How many "hops" away from the starting entity (1-4 levels configurable)
  • Legend: Color-coded legend showing all entity types and relationship types

4. Interact with the Graph

  • Right-click nodes: Context menu with two options:
    • 📋 View Details - Load relationships and highlight Details pane
    • 🎯 Focus on This Node - Enter/exit focus mode for this node
  • Focus Mode: Filters the graph to show only the selected node and its direct neighbors, making it easier to analyze specific connections
  • Drag nodes: Click and drag nodes to reposition them (positions are locked after dragging)
  • Zoom: Use the zoom in/out buttons or scroll wheel
  • Pan: Click and drag the background to move the view
  • Adjust depth: Change max_depth (1-4 levels) to explore deeper relationships
  • Force-directed layout: Graph automatically arranges nodes using physics simulation for optimal visualization

5. Create Relationships

  1. Click "Create Relationship" button
  2. Select source entity (automatically filled if you have one selected)
  3. Choose relationship type
  4. Search and select target entity
  5. Optionally add metadata (JSON format)
  6. Click "Create Relationship"

6. Manage Relationships

In the details panel, you can:

  • View all outgoing relationships (where this entity is the source)
  • View all incoming relationships (where this entity is the target)
  • Delete relationships using the trash icon
Common Use Cases

Track Implementation

Link decisions to the best practices they implement, creating an audit trail of why choices were made.

Decision "Use JWT auth" implements Best Practice "Security guidelines"

Issue Resolution

Connect issues to the work logs and decisions that resolved them.

Work Log "2025-11-18: Fixed auth bug" resolves Issue "Login timeout"

Architecture Dependencies

Map out which architecture components or decisions depend on others.

Architecture "API Gateway" depends_on Architecture "Auth Service"

Pattern Evolution

Show how patterns and practices evolve over time.

Pattern "Microservices v2" supersedes Pattern "Monolith"

Issue Dependencies

Track which issues block others to manage priorities.

Issue "Add OAuth" blocks Issue "Social login feature"
API Integration
Use the REST API to create relationships programmatically

The Knowledge Graph exposes 5 REST API endpoints for full programmatic control:

POST /api/v1/{namespace}/relationships - Create relationship
GET /api/v1/{namespace}/relationships - List with filters
DELETE /api/v1/{namespace}/relationships/{id} - Delete
GET /api/v1/{namespace}/relationships/entity/{type}/{id} - Get related items
GET /api/v1/{namespace}/relationships/traverse/{type}/{id} - Traverse graph
Tips & Best Practices

Start Small

Begin by connecting a few related entities, then expand as patterns emerge

Use Consistent Patterns

Establish conventions for relationship types in your team

Add Metadata

Use the metadata field to add context about why entities are related

Use Focus Mode for Complex Graphs

Right-click a node and select "Focus on This Node" to filter out noise and analyze specific connections

Limit Depth for Performance

Start with depth 2-3 for better visualization performance

Drag to Organize

Drag nodes to organize the graph layout manually - positions are locked after dragging

Clean Up Old Relationships

Delete relationships that are no longer relevant to keep the graph clean