Skip to main content

A personal semantic memory system — store, search, and manage memories locally with vector embeddings

Project description

recall

A personal semantic memory system. Store, search, and manage memories locally using vector embeddings.

Everything runs on your machine: FastAPI server, Qdrant vector database (Docker), and Ollama for local embeddings. Zero cost, full privacy.

Install

From PyPI

pip install recall-cli

From Homebrew

brew tap anelcanto/recall-cli
brew install recall-cli

From source

git clone https://github.com/anelcanto/recall.git
cd recall
./install.sh

Or the quick version:

make install

Prerequisites

  • Docker — runs Qdrant vector database
  • Ollama — local embeddings (brew install ollama && ollama pull nomic-embed-text)
  • uv — Python package manager (brew install uv)

Quick start

# Start services
make up        # Start Qdrant in Docker
make serve     # Start API server (keep this running)

# In another terminal
recall add "The quick brown fox" --tag test
recall search "fox"
recall list
recall status

CLI

recall add "text" --tag work --source cli [--dedupe-key "..."]
recall search "query" --top-k 10 [--no-text] [--output table|json]
recall ingest <file> [--format lines|jsonl] [--source name] [--auto-dedupe]
recall list [--limit 20] [--cursor ...] [--output table|json]
recall delete <id>
recall status

Environment variables

Variable Default Description
RECALL_API_URL http://127.0.0.1:8100 API server URL
RECALL_API_TOKEN (none) Bearer token for auth

Architecture

recall CLI  -->  FastAPI server (:8100)  -->  Qdrant (Docker :6333)
                       |
                       v
                 Ollama (:11434)
                 nomic-embed-text
  • FastAPI serves the HTTP API
  • Qdrant stores vectors and payloads
  • Ollama generates embeddings locally using nomic-embed-text
  • CLI talks to the API over HTTP

User config lives in ~/.memories/.env. Qdrant data persists in a Docker volume.

API endpoints

Method Path Description
POST /memory Store a memory
POST /search Semantic search
POST /ingest Batch import
GET /memories List with pagination
DELETE /memory/{id} Delete a memory
GET /health Service health check

Development

make test              # Unit tests (no services needed)
make test-integration  # Integration tests (Qdrant + Ollama required)
make test-degraded     # Degraded mode tests (Qdrant only)
make test-all          # All tests

License

MIT

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

recall_cli-0.1.0.tar.gz (93.7 kB view details)

Uploaded Source

Built Distribution

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

recall_cli-0.1.0-py3-none-any.whl (16.3 kB view details)

Uploaded Python 3

File details

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

File metadata

  • Download URL: recall_cli-0.1.0.tar.gz
  • Upload date:
  • Size: 93.7 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.6 {"installer":{"name":"uv","version":"0.10.6","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for recall_cli-0.1.0.tar.gz
Algorithm Hash digest
SHA256 7dfd56df9bf3a040d53e00fff3be4bbbbefdb811633fd3d4aa30c3043d957a1c
MD5 dbb5de2b6094dde81522f4ba8f1f687c
BLAKE2b-256 024f031fbbc0227a593c26be958706ee0c6919b1bf5f57f83565004d07eb2720

See more details on using hashes here.

File details

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

File metadata

  • Download URL: recall_cli-0.1.0-py3-none-any.whl
  • Upload date:
  • Size: 16.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: uv/0.10.6 {"installer":{"name":"uv","version":"0.10.6","subcommand":["publish"]},"python":null,"implementation":{"name":null,"version":null},"distro":{"name":"macOS","version":null,"id":null,"libc":null},"system":{"name":null,"release":null},"cpu":null,"openssl_version":null,"setuptools_version":null,"rustc_version":null,"ci":null}

File hashes

Hashes for recall_cli-0.1.0-py3-none-any.whl
Algorithm Hash digest
SHA256 41236bb7b8188e99ecc8e11f5e341e77e6ff33db1569f0116e02f198b4788e94
MD5 5c000bb1e97d33e9baa185e54e00e31e
BLAKE2b-256 dfcdf7b107126e08172e7d72e99b714aca11543e652cfc417dcd129e4d6f8014

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