docs: add Hermes Agent memory integration

[prrao87]

What

Adds an integration page for the LanceDB memory plugin for Hermes Agent under Integrations → AI Platforms & Frameworks.

Hermes Agent (Nous Research) exposes a slot for external memory providers; this plugin fills it, giving Hermes durable, semantic recall across sessions with LanceDB as an embedded, local memory store.

Page structure

• What & why / why LanceDB fits — the lexical-recall gap in built-in memory, and why an embedded vector store is a natural fit.
• Install & activate — <Steps> for plugins install → runtime deps → memory setup → verify.
• The four tools — lancedb_recall / remember / read / forget.
• Walkthrough — the spine of the page: teach it a project convention, then remember → recall → read → forget across a fresh session, showing that a paraphrased query still recalls the fact (semantic vs. keyword).
• Retrieval modes, inspect the store, configuration, and a light one-line benchmark pointer to the blog.

Notes for review

• Follows the existing Agno page's shape, voice, and Mintlify components (<Steps>, <Info>, <Tip>, CodeGroup-style blocks), closing on the Enterprise upsell.
• Intentionally a CLI/YAML/transcript integration — no make snippets entries, since there's no Python/TS/Rust SDK surface to generate. The one Python one-liner (inspecting the store) reads a local ~/.hermes path, so it isn't CI-testable.
• Benchmark kept to a single sentence linking out to the blog and harness, rather than reproducing the LongMemEval methodology in-docs.
• mint validate and mint broken-links pass for this page (the two pre-existing tables/* snippet-export warnings are unrelated).

🤖 Generated with Claude Code

[mintlify]

Preview deployment for your docs. Learn more about Mintlify Previews.

Project	Status	Preview	Updated (UTC)
lancedb-bcbb4faf	🟢 Ready	View Preview	Jul 2, 2026, 7:19 PM