brianpetro/obsidian-smart-connections

Description and use

**Brief Summary (2‑3 sentences)**
*Obsidian Smart Connections* lets you converse with your Obsidian notes and instantly surface AI‑generated links to related content using vector embeddings. It works with local models or any of the 100+ supported APIs (Claude, Gemini, ChatGPT, Llama 3, etc.), turning a personal knowledge vault into a searchable, assistant‑ready knowledge base.

**Value**

• **Searchable internal knowledge** – By embedding every note, the plugin creates a dense, semantic index that goes far beyond keyword search, making it easy for LLM‑driven assistants to retrieve the right context.
• **AI‑augmented navigation** – The chat UI surfaces relevant notes, backlinks, and summaries on‑the‑fly, helping users discover connections they might otherwise miss.
• **Vendor flexibility** – You can run embeddings locally for privacy‑sensitive data or plug into any of the 100+ hosted LLM providers, avoiding lock‑in and allowing cost optimisation.

**Practical Adoption Path**

1. **Install the plugin** in Obsidian (simple UI‑based install).
2. **Choose an embedding source** – either a local model (e.g., sentence‑transformers) for on‑prem security or an API key for a cloud provider you already use.
3. **Run the initial indexing**; the plugin streams notes through the selected model and stores vectors in a local SQLite/LMDB store.
4. **Configure the chat endpoint** (Claude, Gemini, ChatGPT, Llama 3, etc.) and set any rate‑limit or cost controls.
5. **Start chatting** – the assistant will retrieve the top‑k relevant notes via cosine similarity and feed them into the LLM prompt, delivering context‑aware answers and link suggestions.
6. **Iterate** – adjust chunk size, similarity thresholds, or add custom metadata (tags, folders) to fine‑tune relevance for your organization’s use cases (e.g., product documentation, R&D notes, compliance manuals).

**Production‑Readiness**

• **Activity & community** – 4,986 stars, 310 forks, recent commits (as of 2026‑05‑14) and an active issue/PR flow indicate a healthy open‑source project.
• **Maturity** – The plugin already exposes clear integration points (API/SDK/CLI), supports multiple languages, and ships with sensible defaults for both local and cloud embeddings.
• **Scalability** – Vector storage is local but can be swapped for an external vector DB (e.g., Pinecone, Weaviate) without code changes, enabling enterprise‑scale deployments.
• **Risk considerations** – No obvious metadata or licensing red flags, but a final security audit (dependency scanning, supply‑chain review) and confirmation of maintainer responsiveness are advisable before mission‑critical rollout.

Overall, *Obsidian Smart Connections* is a production‑ready OSS candidate for organizations that need to turn their internal note collections into an AI‑searchable knowledge base, with a low‑friction adoption path and flexibility to meet both privacy and cost constraints.

Use in a workflow

brianpetro/obsidian-smart-connections helps make internal knowledge searchable and usable by assistants.

• index knowledge bases
• improve search over documents
• ground assistant answers

Limitations and review notes

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.