Mnemosyne memory provider for Hermes Agent — local-first AI memory with SQLite, vector search, FTS5 hybrid ranking, and episodic consolidation.
Reason this release was yanked:
version reset to 0.1.0
Project description
Mnemosyne for Hermes Agent
Local-first AI memory provider for Hermes Agent.
Powered by Mnemosyne — SQLite with vector search, FTS5 hybrid ranking, episodic consolidation, temporal knowledge graph, and multi-agent validation. Zero cloud. Zero latency. MIT licensed.
Quick Start
pip install mnemosyne-hermes
hermes memory setup # select "mnemosyne"
# Or manually:
hermes config set memory.provider mnemosyne
Why Mnemosyne
- Local-first. Your memory lives on your machine. No cloud. No API key. No network calls.
- 19 tools.
mnemosyne_remember,mnemosyne_recall,mnemosyne_sleep,mnemosyne_validate,mnemosyne_graph_query, and more. - Hybrid search. Vector similarity + FTS5 full-text + temporal scoring. Tunable per-query.
- Episodic consolidation.
mnemosyne_sleepcompresses short-term working memory into long-term episodic summaries. - Knowledge graph.
mnemosyne_triple_addandmnemosyne_triple_queryfor structured fact storage. - Graph traversal.
mnemosyne_graph_queryruns multi-hop BFS through linked memories. - Collaborative validation.
mnemosyne_validatelets agents attest, update, or invalidate each other's memories. - Cross-agent surface.
mnemosyne_shared_rememberstores compact metadata visible across agents.
Configuration
No required config. Defaults use ~/.mnemosyne/ for storage. Optional environment variables:
| Variable | Default | Description |
|---|---|---|
MNEMOSYNE_HOME |
~/.mnemosyne |
Storage directory |
MNEMOSYNE_DB_PATH |
auto | Custom SQLite path |
MNEMOSYNE_VEC_WEIGHT |
0.5 | Vector similarity weight |
MNEMOSYNE_FTS_WEIGHT |
0.3 | Full-text search weight |
MNEMOSYNE_IMPORTANCE_WEIGHT |
0.2 | Importance score weight |
MNEMOSYNE_AUTO_SLEEP_ENABLED |
false | Auto-consolidate after N turns |
MNEMOSYNE_AUTO_SLEEP_THRESHOLD |
50 | Turns between auto-consolidation |
MNEMOSYNE_PROFILE_ISOLATION |
false | Separate DB per Hermes profile |
Links
- Mnemosyne GitHub — core library, benchmarks, docs
- Hermes Agent Memory Providers — provider comparison
- Hermes Memory Provider Plugins — developer guide
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file mnemosyne_hermes-3.1.0.tar.gz.
File metadata
- Download URL: mnemosyne_hermes-3.1.0.tar.gz
- Upload date:
- Size: 34.5 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.4
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
eb7f58e8ca918144f4443cdedc928ab92d6a7bf9c389f60e07b084c691a187b9
|
|
| MD5 |
0495f0c285e45e76e1ee7decc26f39d9
|
|
| BLAKE2b-256 |
e5dd49718981fb38dbafa471b313d0433d4315d7749109a5d951dcf54c7cfc85
|
File details
Details for the file mnemosyne_hermes-3.1.0-py3-none-any.whl.
File metadata
- Download URL: mnemosyne_hermes-3.1.0-py3-none-any.whl
- Upload date:
- Size: 36.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.4
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
8ad831b437ad537fd376a990604fd26bf6b4503229880cb5b9bc967ef3fa9828
|
|
| MD5 |
276e2bd34e206e0c886180d0bd611d1a
|
|
| BLAKE2b-256 |
0b261552fb559bac9464b1ea285138a25356be7dda187f7dcdf5b5e69e6028d9
|