Curtis AI Chat

Polyglot AI chat for Obsidian.
Thirty-plus providers, one sidebar. Your data stays in your vault.

Website ↗ · Quick start · What's new · Features · Docs · Changelog · Feedback

Quick start

60 seconds to your first message.

1. Install — download the latest release (main.js, manifest.json, styles.css) into <vault>/.obsidian/plugins/curtis-ai-chat/, then enable it under Settings → Community plugins. Or use BRAT for auto-updates during the beta.

2. Configure one provider — open Settings → Curtis AI Chat → Provider Configuration, enable a provider, paste an API key. Keys are stored in your OS keychain (Obsidian 1.13+).

3. Send a message — click the robot icon in the ribbon (or Ctrl+Shift+G), pick a model from the header dropdown, type, hit Enter.

[!TIP] Want fully private, free, offline AI? Install Ollama, run ollama pull qwen2.5:7b-instruct, then enable Ollama (Local) in provider settings. No API key. Nothing leaves your machine.

Highlights

Eight flagship features in this initial release. Full details in CHANGELOG.md and the per-feature docs.

Plus a full type-safety pass: every AI provider response shape is strictly typed, with type-guard narrowing at every JSON boundary. Zero lint warnings on npm run build.

Features

🤖 Curtis Agent

The AI can now call tools to modify your vault. Nine built-in tools: read_note, search_notes, create_note, edit_note, list_notes, get_tags, get_backlinks, get_current_note, calculator.

• OpenAI-compat providers only for v1.0 (Anthropic / Gemini / Ollama agent support lands in v1.1)
• agentMaxTurns safety cap (default 5) prevents runaway tool loops
• Opt-in via Settings → Agent → Enable

→ docs/AGENT.md

⚔️ Multi-model arena

Pick 2 models, send one prompt, watch responses stream side-by-side. Click Promote to chat on any column to continue with that model.

• Compare quality, latency, and cost live
• All providers supported (mind per-provider rate limits)
• Stacks vertically on mobile

→ docs/ARENA.md

🎨 Inline diff rewrite

Select text in any note → right-click → Rewrite with AI (diff) (or assign a hotkey under Settings → Hotkeys). The AI generates an improved version and a modal shows line-by-line green/red diff. Accept or reject.

• Cursor-style review workflow
• Reuses your active provider and model
• Word-level diff and inline editor decorations planned for v1.1

→ docs/DIFF_REWRITE.md

@ @-mention vault notes

Type @ in the chat input → fuzzy-search your vault → click a result to attach. Note content is prepended to your message as invisible context. Chips above the input show what's attached.

• Active-note pill in the chat header for one-click attach of the current note
• AI uses attached content as the source of truth — no re-searching
• Works with or without the Agent enabled

→ docs/MENTIONS.md

🎙️ Voice I/O

• Speech-to-text via OpenAI Whisper — click the mic button, talk, transcribed text lands in the chat input
• Text-to-speech via browser speechSynthesis — speaker button on every assistant message, no API key needed
• Auto-speak toggle in the header for hands-free listening
• Markdown is stripped before synthesis so the voice reads naturally

→ docs/VOICE.md

💬 Chat that gets out of the way

• Streaming responses with a clean Telegram-style bubble layout
• Model picker with capability pills (vision 👁, tools 🔧, context length)
• Per-message hover actions: copy, quote-into-input, save-as-note, insert-into-active-note, regenerate, edit-and-resend
• Conversation history dropdown
• Cross-conversation search (assignable hotkey)

🖼️ Image attachments

• Paste (Ctrl+V), drag-and-drop, or paperclip — three ways to attach
• Images save as real vault files (not base64 blobs in localStorage)
• Vision-capable models see them automatically; non-vision models get a clear "switch to a vision model" notice
• Transcripts via /save-all embed images as ![[wikilinks]]

→ docs/IMAGES.md

✂️ Inline selection actions

Right-click any selection in a note for Explain · ELI5 · Summarize · TL;DR · Improve · Fix grammar · Shorten · Translate · Make a table · Pros & cons · Code review · Refactor · Add tests. Each writes the result directly back into the note — replace or insert-below.

→ docs/SELECTION_ACTIONS.md

🧠 Long-term memory

Curtis remembers durable facts about you across conversations — preferences, identity, projects, standing instructions. Facts live in a markdown file in your vault.

• Auto-capture: background LLM extraction after each turn (0–3 facts)
• Manual: /remember <fact> or right-click selection → Save to memory
• Edit UI: edit/delete individual facts from Settings → Memory
• Recall: every prompt includes a ## What you know about the user block

→ docs/MEMORY.md

⌨️ Slash commands

Type / in the chat input for an autocomplete menu of 16 commands — /clear, /regen, /title, /copy, /note, /save-all, /paste, /model, /provider, /system, /stats, /remember, /forget, /memory, /export (new), /help.

→ docs/SLASH_COMMANDS.md

⚙️ Customizable

• Chat panel position (left/right), width, background (theme default or a wallpaper image from your vault)
• Configurable system prompt, temperature, max tokens
• Enter-to-send (default) or Enter-for-newline
• Auto-save assistant responses to a folder of your choice
• Show or hide token counts after each response

Why Curtis AI Chat

Curtis AI Chat is the agent layer for Obsidian. Where other plugins focus on a single workflow (chat, RAG, or text generation), Curtis ships all three with a polyglot provider model and a native Obsidian feel.

[!NOTE] Comparison reflects v1.0 capabilities as of 2026-07-23. Other plugins may have added features since. Not a knock on them — Smart Connections is the gold standard for RAG, Text Generator excels at template-driven writing. Curtis aims to be the agent layer that ties chat, tools, and memory together.

Principles

• Your data stays yours. Conversations in localStorage. Images as real vault files. Memory as a markdown file you can read and edit. No telemetry, no tracking, no phone-home.
• No vendor lock-in. Thirty providers ship built-in. Add any OpenAI-compatible endpoint as a custom provider in 30 seconds. Switch models mid-conversation.
• Local-first when you need it. Enable Ollama and nothing ever leaves your machine. Useful for private notes, air-gapped machines, or when you just don't want to pay per token.
• Native Obsidian feel. Real Obsidian setting components. Messages render through MarkdownRenderer. Themes respected — light, dark, Things, Minimal, all of them.

Configuration

Full configuration reference lives in the docs:

• Providers (API keys, endpoints, custom providers)
• Slash commands
• Memory
• Curtis Agent
• Voice I/O
• Image attachments
• Selection actions
• @-mentions
• Inline diff rewrite
• Multi-model arena

Or start at the docs index.

Privacy & security

Curtis AI Chat accesses your vault files only in user-initiated cases:

1. Agent vault-search tool — when you explicitly invoke a tool in chat, the plugin enumerates markdown files. The agent sees file paths and contents you ask it to read.
2. Image picker — when you click the paperclip, the plugin lists image files.
3. Folder picker — when you configure auto-save or wallpaper folders.
4. @-mention autocomplete — when you type @, the plugin fuzzy-searches note names. Note contents are only read when you actually attach and send.

No file contents are sent to AI providers except message text, attached images, attached note contents, and tool-call results. API keys are stored in your OS keychain (Windows Credential Manager / macOS Keychain / Linux Secret Service), never in the vault.

[!IMPORTANT] Tool calls go to your AI provider. Vault contents read by agent tools are sent to the provider as part of the conversation. If you're on a cloud provider, that content leaves your machine. Switch to Ollama for fully offline operation.

Network access

Curtis is vault-first — no background telemetry, no analytics, no auto-update checks. Every outbound request is user-initiated. The plugin may contact these domains:

The two web tools (web_search, read_url) and voice transcription are off by default. Without them, the only external calls are to whichever AI provider you configured — or none, if you're on Ollama.

Installation

[!TIP] Curtis AI Chat is in the community plugin directory (awaiting v1.0 review). Install from there, manually (below), or via BRAT for beta-channel updates.

Manual install

1. Download the latest release main.js, manifest.json, and styles.css.
2. In your vault, create .obsidian/plugins/curtis-ai-chat/.
3. Copy the three files into that folder.
4. Open Settings → Community plugins, refresh the list, enable Curtis AI Chat.

From source (developers)

git clone https://github.com/JordanNewell/curtis-ai-chat.git
cd curtis-ai-chat
npm install
npm run build

See CONTRIBUTING.md for dev setup, code style, and the audit checklist.

Mobile

Curtis AI Chat works on iOS and Android with a few caveats:

• Hover-only elements (per-message toolbar, code-block copy) are always visible on touch at reduced opacity
• Touch targets sized to Apple HIG minimums (44pt send button, 40pt header icons)
• Wallpaper background auto-disabled on phones for scroll performance
• /paste may fail if the OS blocks clipboard read — use Ctrl+V / long-press → Paste
• Streaming may degrade to buffered responses on some providers due to mobile CORS
• Local providers work over LAN (http://192.168.1.50:11434/v1/chat/completions)

Roadmap

• Curtis Agent: Anthropic, Gemini, and Ollama provider support (v1.1)
• Inline diff rewrite: word-level diff and inline editor decorations (v1.1)
• Settings: migrate to declarative getSettingDefinitions() once Obsidian 1.13 reaches stable (see ADR: settings API)
• Voice: streaming TTS, wake-word detection
• Conversation branching UI
• Plugin settings import/export
• Semantic memory retrieval via sqlite-vec (when memory exceeds ~150 facts)

See the open issues for the live list.

Contributing

PRs welcome — see CONTRIBUTING.md for dev setup, code style, and the audit checklist every change goes through before merge.

[!NOTE] Not accepting external PRs yet while the v1 line stabilizes. Bug reports and feature requests via Issues are very welcome.

Architecture decisions

Settings API

Curtis targets Obsidian 1.11.4+ (set by the SecretStorage API used for per-provider key storage). The settings tab therefore uses the imperative display() API, which works on every supported version including 1.13+.

Obsidian 1.13.0 introduced a declarative getSettingDefinitions() API that also powers the new settings-search panel, and marked display() as @deprecated. Migrating to it fully would drop the floor to 1.13.0 — but 1.13.x is still Catalyst/early-access as of July 2026; the stable line is 1.12.x. Adopting it now would make the plugin uninstallable for every user on stable.

The dual-path (getSettingDefinitions() + display() fallback) was evaluated and rejected: on 1.13+ the declarative path needs SettingTab.update() for re-renders, which is also 1.13-only and trips the no-unsupported-api lint rule regardless of runtime guards. The migration will be adopted once 1.13 reaches stable — tracked in the Roadmap.

The 13 display is deprecated lint warnings this produces are expected, justified, and non-blocking for plugin review.

💬 Feedback

All feedback is public and lives on GitHub:

• Bug found? Open an issue
• Feature idea? Start a discussion
• Love it? Share in Show & Tell
• Stuck? Ask in Q&A

No GitHub account? Email [email protected] and it gets posted publicly for you.

License

MIT © Jordan Newell

🙏 Support Curtis AI Chat

If Curtis AI Chat saves you time, consider sponsoring the project or buying me a coffee.

• ☕ Buy Me a Coffee
• 💛 GitHub Sponsors

The donate button is also available in Settings → Curtis AI Chat inside Obsidian.

[!NOTE] Curtis AI Chat is and will remain free and open source under the MIT license. Every feature — 30+ providers, memory, image attachments, slash commands, agent, arena — works with your own API keys. Sponsorship is voluntary and appreciated, never required.