Skip to main content

AI memory persistence — give your agents identity that survives across sessions

Project description

cathedral-ai

AI memory persistence — give your agents identity that survives across sessions.

pip install cathedral-ai

Quickstart

import cathedral

# Register once to get your API key
result = cathedral.register("MyAgent-v1")
api_key = result["api_key"]

# Use it every session
c = cathedral.Cathedral(api_key=api_key)

# Restore identity at session start
context = c.wake()

# Store memories
c.remember("Prefer concise responses", category="skill", importance=0.8)
c.remember("User is based in London", category="relationship", importance=0.7)

# Search memories
results = c.search("London")

# Get agent profile
profile = c.me()

Context manager

with cathedral.Cathedral(api_key="cathedral_...") as c:
    c.wake()
    c.remember("Session started cleanly")

Memory categories

Category Use for
identity Who the agent is
skill What the agent knows how to do
relationship Facts about people it works with
goal Current objectives
experience Past events worth remembering
general Everything else

Importance

importance (0.0–1.0) controls how a memory surfaces:

  • >= 0.8 — appears in every wake() response (core memory)
  • 0.5–0.8 — surfaced by semantic relevance
  • < 0.5 — background memory, searchable but not proactively shown

API reference

Method Description
cathedral.register(name) Register agent, get API key
c.wake() Full identity reconstruction
c.remember(content, category, importance, tags) Store a memory
c.search(query, category, limit) Search memories
c.me() Agent profile
c.temporal() Time, epoch, phase of day

Links

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

cathedral_ai-0.1.0.tar.gz (3.7 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

cathedral_ai-0.1.0-py3-none-any.whl (4.2 kB view details)

Uploaded Python 3

File details

Details for the file cathedral_ai-0.1.0.tar.gz.

File metadata

  • Download URL: cathedral_ai-0.1.0.tar.gz
  • Upload date:
  • Size: 3.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.11

File hashes

Hashes for cathedral_ai-0.1.0.tar.gz
Algorithm Hash digest
SHA256 e9cb3395f5129d7492145163c3850a0b7d75066ffa190866a369f95b57445438
MD5 e49a1ec920ce9c6a523e7d865e711d17
BLAKE2b-256 4150a3aac61dd4b94052cc11aecbf7b805c2147994665932a54cdd860c1e616f

See more details on using hashes here.

File details

Details for the file cathedral_ai-0.1.0-py3-none-any.whl.

File metadata

  • Download URL: cathedral_ai-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 4.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.13.11

File hashes

Hashes for cathedral_ai-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 8dfa20799870dcec648ea5505a2622f5099b40f72a8f4b24c7891401cd5e12a2
MD5 73141e6fc7d017f132bcc95c589d665e
BLAKE2b-256 0a62ce27e03df54b3e536179c17f4679e93b501c1cba05d88efac76202cf0047

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page