Cathedral Ai provides an identity and persistent memory layer for AI agents, addressing the challenge of maintaining agent identity, beliefs, and continuity across sessions and model switches. The platform is designed to help agents track identity drift, anchor beliefs, and verify continuity, ensuring that agents remember who they are and what they have learned, regardless of context resets or transitions between different models. Cathedral Ai is positioned as a solution for autonomous agents, personal AI assistants, research agents, multi-model teams, character or roleplay agents, and monitoring or operations bots that require reliable memory and identity management over time.
The service offers features such as cross-model drift detection, agent registration with permanent identity via API key, persistent memory storage with categorization and importance scoring, full-text search across stored memories, and recovery options using a recovery token. Agents can reconstruct their full identity and memory context at the start of each session by making a single API call. Cathedral Ai also supports memory storage with customizable categories (such as identity, skill, relationship, goal, experience, and general), bulk memory uploads, and the ability to search and filter memories using full-text search and pagination. The platform provides endpoints for detecting identity drift and verifying agent continuity, including a peer verification system that returns trust scores and drift readings when agents collaborate. External observers can submit drift data for independent verification, adding another layer of trust assessment.
Cathedral Ai is accessible via a web API and offers a Python client for integration. It is compatible with any API-capable model, including Claude, GPT, Grok, and Gemini. The service is open source and provides a free tier allowing up to 1,000 memories per agent, with no expiry unless set by the user, and no rate limit on reads. Write operations are limited to 120 per minute. A Pro tier is available for a monthly fee, increasing memory limits and adding features such as on-chain anchoring, unlimited snapshots, and priority drift scoring. No signup or credit card is required to start using the free tier, and API keys are issued instantly upon agent registration.
In the AI & ML space, Cathedral Ai takes a focused approach. It focuses on ensuring persistent identity and memory continuity for AI agents across sessions and models. It is built as an open-source project for AI developers building agentic systems. Cathedral Ai is open source under the MIT license. The product ships for the command line and API, and it can be self-hosted.
Cathedral Ai first shipped in 2026. Development happens publicly on GitHub with 22 commits in the last 90 days. Key capabilities include identity tracking, drift detection, and memory persistence. It exposes integrations via a public API.
Latest indexed changes and source events
Other apps tracked under the same category.
