All plugins

loredocs

LoreDocs knowledge vaults over MCP: store, version, tag and search project documents, and inject them into agent context. Agent Plugins v1 package (plugin.json + mcp.json + skills/), configuration only: Hermes runs `uvx loredocs==0.1.33` as a local stdio MCP server. Disclosure — runs the loredocs package (BUSL-1.1, with lancedb, fastembed and onnxruntime) via `uvx loredocs==0.1.33` as a local stdio MCP server; stores documents under ~/.loredocs (or LOREDOCS_ROOT), shared with other LoreDocs clients; its add-document tool can read any local file path the agent passes into a vault; no telemetry; a paid Pro license enables semantic search, which downloads a Hugging Face embedding model once.

labyrinth-analyticsToolsv0.1.33Updated 2 hours ago

About

LoreDocs knowledge vaults over MCP: store, version, tag and search project documents, and inject them into agent context. Agent Plugins v1 package (plugin.json + mcp.json + skills/), configuration only: Hermes runs `uvx loredocs==0.1.33` as a local stdio MCP server. Disclosure — runs the loredocs package (BUSL-1.1, with lancedb, fastembed and onnxruntime) via `uvx loredocs==0.1.33` as a local stdio MCP server; stores documents under ~/.loredocs (or LOREDOCS_ROOT), shared with other LoreDocs clients; its add-document tool can read any local file path the agent passes into a vault; no telemetry; a paid Pro license enables semantic search, which downloads a Hugging Face embedding model once.

LoreDocs

LoreDocs is a searchable, organized, version-tracked knowledge base for your AI
projects. Store documents, tag them, search across them, and inject context
into any agent session.

What this plugin installs

Hermes runs uvx loredocs==0.1.33 as a local stdio MCP server. The package
includes 51 MCP tools for vault management, document CRUD, full-text
and semantic search, context injection, and import/export.

Storage

All data is stored locally at ~/.loredocs/ (or override with the
LOREDOCS_ROOT environment variable). Nothing is sent to any cloud service.

Pro license

A Pro license ($9/month) unlocks semantic search, auto-discovered document
relationships, and other Pro-only tools. Semantic search downloads a Hugging
Face embedding model (BGE-small-en-v1.5, ~130 MB) once on first use. The
server reads its own durable license store at ~/.loredocs/license.json --
this is LoreDocs' own data, not another tool's credentials.

If LOREDOCS_PRO is not set, LoreDocs runs on the free tier (limited vaults
and documents).

No telemetry

LoreDocs does not send any telemetry or analytics. All processing is local.

License

The loredocs package is licensed under BUSL-1.1 (Business Source License 1.1).

Long tool calls

By default Hermes reaches MCP tools through its tool-search bridge. With some
models, a call carrying long multi-line text, such as a document body for
vault_add_doc, fails with tool_call 'calls' is not valid JSON (Hermes
issues #119640 and #122711). The model may then retry with shorter, one-line
content, so the document is saved with less detail. To avoid this, turn tool
search off in ~/.hermes/config.yaml:

tools:
  tool_search:
    enabled: "off"

With tool search off, Hermes sends every tool definition to the model on each
turn, which uses more of the context window.

README from the reviewed commit 1f15568; it updates when the author re-pins.

More by labyrinth-analytics

2 plugins by labyrinth-analytics in the catalog.

labyrinth-analyticsloreconvoLoreConvo memory provider: cross-surface persistent session memory with search, recall and structured memory items. Disclosure — stores end-of-session transcripts (truncated, including tool output) and Hermes memory writes in ~/.loreconvo/sessions.db, a store shared across all Hermes profiles and other LoreConvo clients, and injects context from those sessions into each turn; when LoreDocs is installed its session_link_doc tool writes session-to-document links into the LoreDocs database; a Pro license enables a one-time Hugging Face embedding-model download; installs the loreconvo package (BUSL-1.1) with lancedb, fastembed and onnxruntime into the Hermes environment.Communitymemory12

Developer Mode

Explore built-in Hermes toolsets and actions.

HermesAgent

Developer?

Create a Hermes extension for Desktop or agent workflows.

Developer Section