mnemosyne-oss/mnemosyne is a local AI memory layer that stores agent memories in SQLite instead of relying on cloud services. It is intended for coding agents and other agent frameworks, and can be used through MCP or a Python SDK. The catalogue skills connect agents to Mnemosyne’s memory features and related workflows.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add mnemosyne-oss/mnemosyne --skill hermes-memory-providersgit clone --depth 1 https://github.com/mnemosyne-oss/mnemosyneWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/mnemosyne-oss/mnemosyne/hermes-memory-providers)<a href="https://agentmods.dev/skills/mnemosyne-oss/mnemosyne/hermes-memory-providers"><img src="https://agentmods.dev/badge/skills/mnemosyne-oss/mnemosyne/hermes-memory-providers/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/mnemosyne-oss/mnemosyne/hermes-memory-providers"><img src="https://agentmods.dev/badge/skills/mnemosyne-oss/mnemosyne/hermes-memory-providers.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00034 | $0.01807 |
| Opus 5 | $0.00017 | $0.00903 |
| Sonnet 5 | $0.00007 | $0.00361 |
| Haiku 4.5 | $0.00003 | $0.00181 |
Grade A, and why
hermes-memory-providers scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 213 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Mnemosyne — Hermes Memory Provider
Mnemosyne is a local-first memory layer for AI agents. When deployed as a Hermes memory provider, it replaces the built-in MEMORY.md/USER.md system with SQLite-backed vector + FTS5 hybrid search, episodic consolidation, temporal knowledge graphs, and optional bidirectional sync.
100% local. Zero cloud. Sub-millisecond recall.
What It Gives You
- System prompt injection —
# Mnemosyne Memorycontext block in every prompt - Pre-turn prefetch — relevant memories injected before each LLM call
- Post-turn sync — conversation turns auto-stored to episodic memory
- 20 tools auto-injected into the model's tool surface (remember, recall, sleep, triples, scratchpad, graph, sync, diagnostics, etc.)
- 3 lifecycle hooks —
pre_llm_call,on_session_start,post_tool_call - CLI commands —
hermes mnemosyne {stats|sleep|inspect|export|import|clear|version}
All without touching Hermes core — deployed purely through the plugin directory.
Quick Check
hermes memory status # See active provider and installed plugins
Install
Step 1 — Install the package
pip install mnemosyne-hermes
Debian/Trixie users (bare pip blocked): use a venv first:
python3 -m venv ~/.hermes/hermes-agent/venv
source ~/.hermes/hermes-agent/venv/bin/activate
pip install mnemosyne-hermes
mnemosyne-hermes wraps the core mnemosyne-memory library with the plugin
manifest and entry points Hermes needs. It does not pull embeddings or LLM
deps — pair it with one of:
| Extra | When | RAM |
|---|---|---|
| (core only) | Raspberry Pi, remote embedding API | ~50 MB |
mnemosyne-memory[embeddings] |
Local vector search (fastembed ONNX) | ~800 MB |
mnemosyne-memory[all] |
Local embeddings + local LLM consolidation | ~1.5 GB |
Step 2 — Link the plugin
mnemosyne-hermes install
This creates the symlink ~/.hermes/plugins/mnemosyne/ → <installed package>
so Hermes discovers it on startup.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 213 lines · 34 tokens per session scan A c8f631d4c543
hermes-memory-providers is a skill published in the GitHub repository mnemosyne-oss/mnemosyne (3,093 stars, last pushed today), licensed MIT. It adds 34 tokens to every session and 1,807 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
honcho
Configure and use Honcho memory with Hermes -- cross-session user modeling, multi-profile peer isolation, observation config, dialectic reasoning, session summaries, and context budget enforcement. Use when setting up Honcho, troubleshooting memory, managing profiles with Honcho peers, or tuning observation, recall…
qmd
Search personal knowledge bases, notes, docs, and meeting transcripts locally using qmd — a hybrid retrieval engine with BM25, vector search, and LLM reranking. Supports CLI and MCP integration.
llm-wiki
Karpathy's LLM Wiki: build/query interlinked markdown KB.
autocontext-consumer
Use when an agent needs to USE knowledge Autocontext already produced - find which scenarios have knowledge, read the playbook and lessons for one, understand the on-disk file and folder layout, and move knowledge between checkouts. Host-agnostic; requires only the autoctx CLI and the filesystem.
wikiskill-maintainer
Consolidate traces into the persistent wiki (WikiSkill).
mnemosyne-maintenance
Use when: upgrading Mnemosyne, diagnosing slow/hung consolidation (mnemosynesleep), fixing missing embeddings, or troubleshooting import/version mismatches.