Smart Composer
Plugin : A set of capabilities installed into an AI application or a host · requires obsidian
Cursor-style assisted editing in Obsidian, with vault context and local model support.
Promising, still in beta.
Security reflects where data goes, requested permissions and incidents on record at the verification date. It is not a guarantee of zero risk: check the deployment mode and the connections you enable.
What it does, concretely
Every time you query an AI assistant, you have to provide all the background context from scratch. Smart Composer solves this by embedding AI directly into Obsidian, where your notes already live, so you no longer need to copy and paste your background information.
In practice, you open a chat panel inside Obsidian and mention files or folders using the @filename syntax. The plugin sends that content as context to the AI, which can then suggest edits directly in your note. You accept or reject each suggestion with a single click. A semantic search across your entire vault (RAG) can also automatically surface the most relevant notes.
Smart Composer is aimed at Obsidian users who regularly write long-form content and want to use AI without leaving their workspace or compromising data privacy, thanks in particular to support for local models via Ollama.
However, the plugin is maintained by a single developer who has stated it is no longer under active development. Users who need responsive support or a guaranteed roadmap may find this situation concerning.
What it offers
- ·Contextual chat: mention files or folders with @filename to include them precisely in the conversation.
- ·Apply edits: the AI suggests changes to your note, which you accept or reject with a single click.
- ·Vault semantic search (RAG): automatically find relevant notes using Cmd+Shift+Enter.
- ·Multimedia context: add web links, images (drag and drop, clipboard) and YouTube videos whose transcripts are fetched automatically.
- ·MCP support: connect Smart Composer to external MCP servers to access third-party tools and data sources.
- ·Local models: use Ollama or LM Studio to run LLMs locally without sending your data to a remote server.
- ·Prompt templates: create reusable templates accessible by typing / in the chat, to standardize repetitive tasks.
- ·Direct connection to OpenAI and Gemini subscriptions without an API key (use with caution for Claude, account bans have been reported).
In practice
- Price
- Freemium
- Commercial use
- Commercial use: yes
- Available on
- obsidian
- Host
- obsidian
- Licence
- MIT
- FR interface
- No
Security and data
- Rating
- Security: one caveat, see the card
- Your data
- Local or cloud depending on use
- Maintenance
- Inactive for over six months
- Last release
- 26 January 2026
Links and repository checked on 25 October 2026 · added on 1 November 2024