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
dax8it/total-recall
Total Recall is a framework-agnostic continuity memory engine that treats provenance rather than recall as the hard problem: it aims to let an agent prove that the memory it is about to use is still authoritative. An append-only event ledger with hash chaining is the source of truth; state is derived from it by deterministic reduction, captured in signed checkpoints and anchors, and verified fail-closed before rehydration. On top of that trust spine sit cited recall, a knowledge engine, freshness reporting that flags stale or superseded promises and decisions, temporal graph timelines answering what was known at a given moment, and explicit federation between workspaces instead of implicit cross-workspace sharing. Retrieval indexes over LanceDB, QMD and SQLite full-text search are derived and rebuildable, with a lexical fallback. Operational features include encrypted backup push and pull, device identity, single-writer leases, fork-import quarantine, receipt verification, handoff artifacts, incident records, and an operator console covering trust, incidents, knowledge, vault and backups. The engine itself depends on no particular agent runtime. Hermes Agent integration is optional and ships as a memory-provider plugin under a hermes-plugin directory. The installer detects the Python environment Hermes actually runs under, installs or upgrades total-recall-core there, writes the provider bundle into the Hermes plugins path together with a flat compatibility path used by Hermes v0.15.x, validates the bundle, selects total-recall as the memory provider for the chosen profile, and writes matching context compression and auto-rehydrate defaults. A doctor command reports readiness and a bundle command produces a distributable archive. Other ingestion paths cover local files and folders plus working-context sources such as meetings, email, chat, tickets and calendars, with Obsidian vault export and an explicit preview-and-promote import. Licensed AGPL-3.0-or-later, with a separate commercial licence for uses that cannot comply.
https://github.com/dax8it/total-recall
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.