Skip to main content

AI KNOW - Memory Kernel for long-running AI agents (Hermes ecosystem)

Project description

AI KNOW - Python MVP

Python 3.11 + FastAPI implementation of the AI KNOW Memory Kernel v0.1. This is the primary execution path for v0.1; the Rust code under src/backend/memory-core/ is retained as a reference implementation.

Layout

python-mvp/
├── ai_know/
│   ├── main.py          # FastAPI app factory
│   ├── config.py        # Pydantic Settings (env-driven)
│   ├── api/             # HTTP routers (health, events, memories, ...)
│   ├── core/            # Ingestor, extractor, arbitrator, ...   (PR 3+)
│   ├── db/              # SQLAlchemy models + repositories       (PR 2)
│   ├── schemas/         # Pydantic request/response schemas       (PR 3+)
│   ├── providers/       # Embedding + LLM providers               (PR 4+)
│   └── jobs/            # Background jobs (dream, consolidation)  (PR 7+)
├── migrations/          # Alembic                                  (PR 2)
├── tests/
│   ├── unit/            # No I/O
│   ├── integration/     # Needs Postgres / Redis                    (PR 2+)
│   └── evals/           # Mini evals                                 (PR 8)
├── requirements.txt
├── pyproject.toml
├── Dockerfile
└── README.md

Quick start

From the ai-know repo root:

docker compose up --build
# One-shot: apply the schema.
docker compose exec api alembic upgrade head
curl http://localhost:8090/health
# → {"status":"ok","version":"0.1.0","environment":"dev"}
curl http://localhost:8090/health/ready
# → {"status":"ok",...,"database":"ok","redis":"ok"}

Host ports: API 8090, Postgres 5433, Redis 6380 (they coexist with Hermes on 8080/5432 and Claudio on 3001).

Local dev (no Docker)

cd src/backend/python-mvp
python3.11 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
export DATABASE_URL=postgresql+psycopg://ai_know:ai_know_dev@localhost:5433/ai_know
export REDIS_URL=redis://localhost:6380/0
alembic upgrade head
uvicorn ai_know.main:app --reload --port 8080

Tests

Unit only (no I/O):

pytest tests/unit -v

Integration (needs docker compose up):

# From inside the running api container:
docker compose exec api python -m pytest tests/integration -v

PR status (as of v0.1.0 tag)

PR Scope Status Tests
1 scaffold / FastAPI / health ✅ done
2 DB migrations (9 tables + repos) ✅ done 14/14
3 Events API ✅ done 44/44
4 Episode + Claim extraction ✅ done 82/82
5 Arbitration + consolidation ✅ done 109/109
6 Context Pack builder ✅ done 149/149
7 Feedback engine ✅ done 174/174
8 Mini evals ✅ done 178/178
9 Hermes adapter (dogfooding) ✅ done 187/187
10 v0.1 release polish ✅ done 187/187

v0.1 backend RELEASE-READY. See the top-level CHANGELOG.md for landing dates and commits. See ai_know/adapters/hermes.py for the client-side SDK that lets hermes-agent (and any other host) talk to this backend without reimplementing the wire contract.

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

hermes_ai_know-0.3.0.tar.gz (94.1 kB view details)

Uploaded Source

Built Distribution

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

hermes_ai_know-0.3.0-py3-none-any.whl (124.4 kB view details)

Uploaded Python 3

File details

Details for the file hermes_ai_know-0.3.0.tar.gz.

File metadata

  • Download URL: hermes_ai_know-0.3.0.tar.gz
  • Upload date:
  • Size: 94.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for hermes_ai_know-0.3.0.tar.gz
Algorithm Hash digest
SHA256 708d6982b3ec6306b0a46102bcf693f9692b7dace76d75ab23ea7603cf8ffcb5
MD5 01d1447dd310588c75554591fc5cfeed
BLAKE2b-256 93eebb3cc374cd2b55c14e0e13c023f8db317c35e3f8c68662d66dc7f6d23c60

See more details on using hashes here.

File details

Details for the file hermes_ai_know-0.3.0-py3-none-any.whl.

File metadata

  • Download URL: hermes_ai_know-0.3.0-py3-none-any.whl
  • Upload date:
  • Size: 124.4 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for hermes_ai_know-0.3.0-py3-none-any.whl
Algorithm Hash digest
SHA256 a8ad48fffc74daf3807d8d0e463967de1067708d566e7808ea37409c3361d0a6
MD5 ca58069c552c5e6d6bd4bd74f31ed378
BLAKE2b-256 5f48f72182c32239a0b7711810459384891c708b2eee65d5a7e4641dc193671a

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