Skip to main content
ACE3 LogoACE3
Public BetaWe're in Public Beta. Things may break. Please report issues via the Support tab.

The Complete AI
Memory Platform

9 entity types, knowledge graph, MCP server, AI builder, semantic search - everything you need for production AI memory

Structured Memory

9 Specialized Entity Types

Not just key-value storage. ACE provides structured entities designed for how developers actually work.

Issues

Track bugs, features, technical debt with severity, priority, error messages, stack traces, reproduction steps

Decisions

Architecture Decision Records (ADR) with rationale, alternatives considered, consequences, and status

Work Logs

Daily progress tracking with tasks completed, hours worked, context, and linked issues/decisions

Architecture

System components across layers (frontend, backend, database, API) with technology choices and trade-offs

Best Practices

Team coding standards, guidelines, and conventions with categories and code examples

Patterns

Reusable code templates with problem-solution mapping, use cases, pros/cons

Tags

Cross-entity labels for organization with custom colors and descriptions

Relationships

Knowledge graph links between any entities with 12 semantic relationship types

Generic Memories

Flexible storage for anything else with title, content, category, and custom metadata

Knowledge Graph

12 Relationship Types

Connect any entity to any other. Your AI traverses the graph to discover context automatically.

Relationship Types
implementsreferencessupersedesrelates_tocaused_byresolvesdocumented_independs_onblocksduplicatesvalidatesreproduces
Graph Traversal

Breadth-first search with configurable depth and direction. Find related context your AI needs.

  • Decision - Issues it caused - Solutions that fixed them
  • Bug - Root cause - Prevention pattern
  • Architecture - Best practices - Code patterns
MCP Server

21 Native AI Tools

Model Context Protocol gives your AI direct access to ACE. Works with Claude Code, Codex CLI, Gemini CLI, Cursor, Windsurf, ChatGPT, and any MCP client.

Memory

Store and retrieve with semantic search

ace_remember
ace_recall
ace_get_context

Issues

Full issue lifecycle management

ace_track_issue
ace_resolve_issue
ace_find_solutions

Knowledge

Capture decisions and patterns

ace_track_decision
ace_architecture
ace_patterns
ace_best_practices

Graph

Navigate relationships

ace_graph_traverse
ace_graph_connect
ace_graph_related
ace_graph_list

Search

Deep search and organization

ace_search
ace_manage_tags

Work

Track progress and metrics

ace_log_work
ace_get_stats

AI Builder

Generate plans from requirements

ace_generate_plan
ace_create_plan

Observer

Proactive quality monitoring

ace_observe

Also Includes

  • 9 read-only resources
  • 6 workflow prompts
  • Full MCP compliance
AI Builder

Generate Plans from Requirements

Describe what you want to build. ACE generates structured entities with relationships - ready to import.

1Describe

Enter your requirements in natural language. Add context about your tech stack and constraints.

2Generate

AI creates decisions, architecture, patterns, best practices, issues - all linked with relationships.

3Import

Review the plan, make edits if needed, then import atomically. Everything linked and searchable.

Works with:
GPT-4ClaudeGemini
REST API

60+ Endpoints

Universal access for any AI. Full CRUD for all entity types plus search, bulk operations, and more.

# Store a memory
curl -X POST \
  http://localhost:7777/api/v1/my-project/memory \
	  -H "Authorization: Bearer $TOKEN" \
  -d '{"key": "auth-pattern", "value": {"pattern": "JWT with refresh tokens"}}'
# Track an issue
curl -X POST \
  http://localhost:7777/api/v1/my-project/issues \
	  -H "Authorization: Bearer $TOKEN" \
  -d '{"title": "Login fails on Safari", "severity": "high", "category": "bug"}'
Semantic Search

Find Context Intelligently

Not keyword matching. True semantic understanding with OpenAI embeddings + pgvector.

Cross-Entity Search

Search across memories, issues, decisions, architecture, patterns - all at once.

Problem-Solution Matching

Find resolved issues similar to your current problem. Reuse past solutions automatically.

Context Retrieval

Get comprehensive context combining memories, issues, decisions, and best practices for any query.

Team Collaboration

Shared AI Memory

Team knowledge that persists. New members inherit context. Decisions documented once, used forever.

Namespaces

Organize by project, team, or purpose. Complete isolation between namespaces.

Role-Based Access

Owner, Admin, Write, Read permissions. Invite by email. Manage from dashboard.

Real-Time Sync

Everyone sees the same context. No stale data. Instant updates across all team members.

Observer

Proactive Quality Monitoring

Catch mistakes before they reach production. Observer reviews AI output against your project memory, checking for errors, contradictions, and alignment issues.

Error Detection

Automatically scan AI-generated code, decisions, or documentation for potential errors, inconsistencies, and anti-patterns.

Memory Alignment

Check new content against existing decisions, patterns, and best practices. Detect contradictions before they cause problems.

Role-Based Analysis

Get feedback from different perspectives - developer, architect, PM, QA, or security. Each role focuses on relevant concerns.

# Review AI output with Observer
ace_observe({
  content: "Generated authentication code...",
  context: "Implementing user login flow",
  role: "security"
})

// Observer checks against your project memory
// and returns warnings, suggestions, and guidance