memento-collector
Cross-platform agent that automatically collects AI coding tool memory files and syncs them to a central Memento server (a shared brain for your AI coding tools).
Supported AI Tools
| Tool | Data Collected |
|---|---|
| Claude Code | Conversations, memory, plans, history |
| OpenClaw | Sessions, identity, memory, learnings, skills |
| Codex | Sessions (active + archived), history, skills |
| Antigravity | Full conversations (built-in .pb decryption — AES-256-GCM + protobuf), brain plans, code snapshots |
| Obsidian | All markdown notes in your vault |
| Cursor | Agent transcripts, skills, MCP config |
Install
pip install memento-brain-collector
Antigravity support (decrypting encrypted .pb conversation files) is built in —
no extras needed. The cryptography library is already a required dependency.
Quick Start
# Interactive setup (first time)
memento-collector setup
# Or run directly
memento-collector run
The setup wizard will:
- Detect your platform (macOS / Linux / Windows)
- Auto-discover installed AI tools and Obsidian vaults
- Configure the server URL and auth token
- Optionally install as a system service (auto-start on boot)
Commands
memento-collector setup # Interactive setup wizard
memento-collector run # Run in foreground
memento-collector install # Install as system service
memento-collector start # Start the service
memento-collector stop # Stop the service
memento-collector status # Show collector status
memento-collector uninstall # Remove system service
How It Works
- File Watching — Uses
watchdog(FSEvents on macOS, inotify on Linux, ReadDirectoryChanges on Windows) to detect file changes in real-time - Parsing — Supports Markdown, JSONL, JSON, TOML, SQLite formats
- Sanitization — Automatically redacts API keys, tokens, passwords, private keys before upload
- Queuing — Local SQLite queue for offline resilience (syncs when server is reachable)
- Syncing — HTTP upload to server, with chunked upload for files > 2MB (tested with 37MB files)
- Device Identity — Each device gets a persistent unique ID, all data tagged with device info
Configuration
Environment variables (or set via memento-collector setup):
| Variable | Default | Description |
|---|---|---|
MEMENTO_SERVER_URL |
http://localhost:8001 |
Server API URL |
MEMENTO_SERVER_TOKEN |
Collector auth token | |
MEMENTO_OBSIDIAN_VAULT_PATH |
Auto-detected | Obsidian vault path |
Config file: ~/.memento/config.json
System Service
| Platform | Service Type | Config Location |
|---|---|---|
| macOS | LaunchAgent | ~/Library/LaunchAgents/com.memento.collector.plist |
| Linux | systemd user | ~/.config/systemd/user/memento-collector.service |
| Windows | Task Scheduler | MementoCollector scheduled task |
License
MIT
Release files for memento-brain-collector 0.0.15
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| memento_brain_collector-0.0.15.tar.gz | 70.6 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| memento_brain_collector-0.0.15-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 154.0 kB
Release files / memento_brain_collector-0.0.15.tar.gz
| Download URL | memento_brain_collector-0.0.15.tar.gz |
|---|---|
| Size | 70.6 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
3975705f5cc1473ac5f2f780cae0b5f6bf47406cc998c7ca84a07bac605ff27b
|
|
BLAKE2b-256 checksum How to use checksums |
4c1f0e920160c7643dc5b57f24197d77c41915a2b48d60baf1627517144e8f21
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 5, 2026.
Transparency logRelease files / memento_brain_collector-0.0.15-py3-none-any.whl
| Download URL | memento_brain_collector-0.0.15-py3-none-any.whl |
|---|---|
| Size | 83.4 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
d0e65b3383b926995685a052a6079cacebbbe4bf683bdb7fd7656069ba6fae53
|
|
BLAKE2b-256 checksum How to use checksums |
ab3a937d5930c685202a5eb30a329c480713ad269ad44c67682b54e43a37cac5
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
Yes |
| Uploaded via |
twine/7.0.0 CPython/3.13.14
|
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
PyPI Publish Attestation
PyPI verified that this artifact, at this checksum, originated from the publisher listed below.
Signed by GitHub Actions, verified by PyPI on Sep 5, 2026.
Transparency log