volcengine/OpenViking
A full bidirectional MemoryProvider plugin bringing OpenViking's viking:// context database — tiered retrieval, automatic session-end extraction, and semantic search — to Hermes Agent.
Awesome Hermes Plugins › Memory Providers
NehuenD/hermes_mempalace
MemPalace is a Hermes Agent memory-provider plugin that combines a ChromaDB vector store for semantic search with a SQLite-backed knowledge graph for structured facts, running entirely locally with no external accounts and no network calls for its core operation. It organizes stored memories using a "palace" taxonomy with four levels: a Wing for a high-level domain (defaulting to wing_general but configurable), a Room for a topic or activity such as learnings, sessions, or preferences, a Closet for a content classification like personal, projects, or world, and a Drawer holding one individual memory entry — a fact, preference, or note carrying a UUID, content, metadata, and an optional time-to-live. An optional compact shorthand dialect called AAAK lets dense entries be stored more tersely. Once installed under ~/.hermes/plugins/mempalace/ (or a named profile's plugins directory) with its chromadb and pyyaml dependencies, it is activated by setting hermes config set memory.provider mempalace, since only one external memory provider can be active at a time and installing the plugin does not by itself enable it. Once active it exposes 45 tools alongside Hermes's built-in ones, including mempalace_remember for auto-classifying and embedding a new memory for later semantic search, mempalace_recall for ranked semantic search across drawers with subject, closet, category, and flag filters plus pagination, and mempalace_learn, which runs an optional duplicate check before filing a fact and, if a near-duplicate already exists, returns that entry instead of creating redundant content, with updates instead creating a new version chained by a parent id. A separate session-tracking pair, mempalace_session_write and mempalace_session_read, lets an agent record an end-of-session summary and next steps for a project and restore that context automatically at the start of a later session, alongside a broader knowledge-graph query surface for structured facts.
https://github.com/NehuenD/hermes_mempalace
Also in Memory Providers
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.
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.