TencentCloud/TencentDB-Agent-Memory
TencentDB Agent Memory is a Hermes Agent plugin providing a 4-tier progressive and local-first memory pipeline to enhance AI agents by reducing token usage and improving task success rates.
Awesome Hermes Plugins › Memory Providers
RohiRIK/OpenLtm
OpenLTM is a persistent semantic memory layer for AI coding agents, released under MIT. The core is a TypeScript package, @rohirik/openltm-core, with thin adapters per host; Claude Code, OpenCode and an experimental Pi adapter are supported alongside a native Hermes Agent integration. The Hermes side is a Python directory plugin living under hermes/ in the repository, installed with hermes plugins install RohiRIK/OpenLtm/hermes/openltm_hermes and selected through memory.provider: openltm_hermes in the Hermes configuration file. It stores everything in a local SQLite database at ~/.hermes/openltm.db with no server and no network calls, and its canonical schema folds earlier migrations so that fresh installs do not crash the learn and janitor code paths. The project ships 38 plugin tests for the Hermes integration. The design rests on four stated principles. Memory should be automatic, so hooks do the work: a session end hook extracts patterns and a session start hook injects them back. Decay is treated as a feature, with stale entries ageing out while anything marked at maximum importance stays permanent. Retrieval is semantic rather than keyword-first: FTS5 full-text search runs first and vector embeddings take over when it returns nothing, so a query phrased differently from the stored text still matches. Configuration is meant to be unnecessary, and the database sits outside the plugin directory so it survives updates. Beyond storage, memories form a graph whose relationships can be traversed for reasoning chains, and a browser-based explorer visualizes the network. Vector recall uses sqlite-vec when available and degrades to a JavaScript cosine implementation when the system libsqlite3 extensions are missing. A memory subcommand exposes learn, recall, forget and context operations to headless shells, cron jobs and CI, and the same core can run as a stdio MCP server for any MCP-capable host.
https://github.com/RohiRIK/OpenLtm
Also in Memory Providers
MemoriLabs/Memori
Memori is an agent-native memory infrastructure, offering an LLM-agnostic layer that converts agent execution and conversation into structured, persistent state for production systems, including a ...
basicmachines-co/basic-memory
Basic Memory is a local-first, knowledge graph-based memory system for AI agents, providing persistent, bidirectional knowledge access via Markdown files and the MCP protocol, including a Hermes pl...
agentscope-ai/ReMe
A local-first Markdown memory service with a Hermes Agent plugin that turns conversations and documents into linked, searchable notes using wikilinks, BM25 and optional embedding retrieval.
mnemosyne-oss/mnemosyne
Mnemosyne is a zero-dependency, sub-millisecond AI memory system for Hermes Agents and other LLM frameworks, focusing on efficient, local, and private knowledge retention.
ClaudioDrews/memory-os
Memory OS is a 7-layer memory operating system for Hermes Agent, providing persistent memory, structured facts, fabric recall, an auto-curated wiki, and surgical context injection, running locally ...
zhangfengcdt/memoir
Memoir is a high-performance semantic memory system for AI agents, offering Git-like version control for AI memory management to combat context contamination, token costs, and memory drift.
MaxFreedomPollard/Compartment
Offline, encrypted agentic memory vault that installs into Hermes Agent as a plugin and MCP server, keeping one transferable memory store shared by every agent on the machine.