Per-Project Notes
Each project has its own notes panel:- Project-specific context — Architecture decisions, API keys, deployment steps
- Session continuity — Document what you learned so the AI remembers next time
- Searchable — Find notes by keyword across all projects

Creating Notes
- Open the Notes tab in the project window
- Click New Note
- Enter a title
- Write content in the editor (supports plain text and markdown-style formatting)
- Click Save
Editing Notes
Click a note in the list to open it in the editor. Make changes and click Save to persist them.Deleting Notes
Click the trash icon in the note editor toolbar. Deleted notes are removed permanently (no undo).Pinning Notes
Pin important notes to keep them at the top of the list:- Open a note
- Click the pin icon in the toolbar
- The note moves to the top and displays a pin badge
Global Notes
In addition to per-project notes, CodeFire supports global notes that span all projects:- Accessible from the home view
- Useful for cross-project knowledge (team conventions, infrastructure notes)
- Searchable alongside project-specific notes
Rich Editor
The note editor supports:- Monospace font — Great for code snippets
- Scrollback — Notes can be arbitrarily long
- Auto-save — Click Save or use Cmd+S
- Markdown-style formatting — Use
#for headings,*for lists, etc. (displayed as plain text, but you can use it for structure)
CodeFire’s editor currently displays plain text. Markdown rendering is planned for a future release.
Search Notes
Use the search bar in the Notes tab to filter by keyword:- Searches titles and content
- Case-insensitive
- Instant filtering as you type
MCP Integration
When the CodeFire MCP server is installed, your AI agent can:- Read notes — Query notes by project or globally
- Search notes — Find notes by keyword
- Create notes — Add new notes programmatically
- Update notes — Append content to existing notes
- You document context in notes
- The AI reads those notes when starting a session
- The AI can update notes with new learnings
Example: Agent Reading Notes
Best Practices
Document architecture decisions
Document architecture decisions
When you make a significant technical decision (e.g., “Use PostgreSQL instead of MongoDB”), create a note explaining why. Future sessions benefit from this context.
Keep API keys in notes (but encrypt them)
Keep API keys in notes (but encrypt them)
Store API keys, tokens, and credentials in notes for easy access. Use a password manager or encryption tool for sensitive values.
Use notes as a knowledge base
Use notes as a knowledge base
Treat notes like a wiki for your project. Document setup steps, common commands, edge cases, and gotchas.
Pin frequently-used notes
Pin frequently-used notes
Pin notes you reference often (e.g., environment variables, test credentials, deployment checklist).
Let the AI update notes
Let the AI update notes
Enable the MCP server and allow the AI to append learnings to notes. This creates a living knowledge base that grows with each session.
Note Metadata
Each note stores:- Title — Short, descriptive name
- Content — Free-form text (supports markdown syntax)
- Pinned — Boolean flag (pinned notes appear first)
- Created — Timestamp
- Updated — Last modification timestamp
- Project ID — Links note to a specific project (or global if null)
Export Notes
Notes are stored in SQLite (~/Library/Application Support/CodeFire/codefire.db). Query directly for export:
Related Features
Project Management
Each project has its own notes panel
Task Tracking
Link notes to specific tasks for context
Session Monitoring
Reference notes during sessions for context
