Add supermemory extension

[MaheshtheDev]

Description

This extension integrates Supermemory's intelligent memory system directly into Raycast, enabling users to instantly capture, organize, and retrieve information without leaving their workflow.

Features

Add Memory Command: Capture text, URLs, articles, or any information directly from Raycast with project organization support
Search Memories Command: Real-time AI-powered search through your entire knowledge base with relevance scoring and rich metadata

Screencast

Checklist

• I read the extension guidelines
• I read the documentation about publishing
• I ran npm run build and tested this distribution build in Raycast
• I checked that files in the assets folder are used by the extension itself
• I checked that assets used by the README are placed outside of the metadata folder

[raycastbot]

Congratulations on your new Raycast extension! 🚀

Due to our current reduced availability, the initial review may take up to 10-15 business days

Once the PR is approved and merged, the extension will be available on our Store.

[greptile-apps]

Greptile Overview

Summary

This PR adds a new Supermemory extension that integrates AI-powered memory management directly into Raycast. The implementation includes two main commands: Add Memory for capturing and organizing information, and Search Memories for AI-powered retrieval with relevance scoring.

Key Changes:

• Complete extension implementation with proper TypeScript types and error handling
• API integration with authentication and structured error responses
• Form-based memory creation with project organization support
• Real-time search with debouncing and rich result presentation
• Proper Raycast UI patterns with loading states and empty views

Minor Issues Found:

• Unnecessary Preferences interface definition (autogenerated by Raycast)
• Code duplication in utility functions for content extraction
• Debug logging statement that should be removed
• Changelog needs descriptive content

The extension follows Raycast development patterns well and provides a clean user experience for memory management workflows.

Confidence Score: 4/5

• This PR is safe to merge with minor cleanup recommended
• The extension is well-implemented with proper error handling, good UX patterns, and follows Raycast conventions. The identified issues are minor style improvements and cleanup items that don't affect functionality.
• Minor attention needed for extensions/supermemory/src/api.ts and extensions/supermemory/src/search-memories.tsx to address code duplication and cleanup

Important Files Changed

File Analysis

Filename	Score	Overview
extensions/supermemory/src/api.ts	3/5	API implementation with good error handling, but includes unnecessary Preferences interface and debug logging
extensions/supermemory/src/add-memory.tsx	5/5	Clean implementation of add memory form with proper validation and loading states
extensions/supermemory/src/search-memories.tsx	3/5	Feature-rich search interface with good UX, but contains code duplication in utility functions

_{9 files reviewed, 4 comments}

_{Edit Code Review Agent Settings | Greptile}

[pernielsentikaer]

@greptileai can you do a fresh review?

[greptile-apps]

I'll do a fresh review of this new Supermemory extension. Let me examine the code structure and implementation.

Overall Assessment

This is a well-structured extension that integrates Supermemory's knowledge base with Raycast. The code follows good practices and provides a clean user experience with two main commands: adding memories and searching through them.

Positive Aspects

✅ Clean Architecture: Well-organized with separate API layer, proper TypeScript types, and modular components
✅ Error Handling: Comprehensive error handling with user-friendly toast messages
✅ User Experience: Debounced search, loading states, and intuitive UI patterns
✅ Security: API key stored securely in preferences
✅ Documentation: Good README with clear setup instructions

Issues Found

1. Package.json Issues

File: extensions/supermemory/package.json

The eslint config reference is incorrect:

{
  "$schema": "https://www.raycast.com/schemas/extension.json",
  "name": "supermemory",
  "title": "Supermemory",
  "description": "Add and search memories with your personal AI-powered knowledge base",
  "icon": "extension-icon.png",
  "author": "supermemory",
  "platforms": [
    "macOS",
    "Windows"
  ],
  "categories": [
    "Productivity",
    "Web"
  ],
  "license": "MIT",
  "commands": [
    {
      "name": "add-memory",
      "title": "Add Memory",
      "subtitle": "add memory to your supermemory app",
      "description": "Add text, URLs, or documents to your supermemory knowledge base",
      "mode": "view"
    },
    {
      "name": "search-memories",
      "title": "Search Memories",
      "description": "Search through your saved memories and find relevant information",
      "mode": "view"
    }
  ],
  "preferences": [
    {
      "name": "apiKey",
      "type": "password",
      "required": true,
      "title": "API Key",
      "description": "Your Supermemory API Key. Get it from https://supermemory.link/raycast",
      "placeholder": "Enter your Supermemory API Key"
    }
  ],
  "dependencies": {
    "@raycast/api": "^1.103.2",
    "@raycast/utils": "^1.17.0"
  },
  "devDependencies": {
    "@raycast/eslint-config": "^1.0.11",
    "@types/node": "22.13.10",
    "@types/react": "19.0.10",
    "eslint": "^8.57.0",
    "prettier": "^3.5.3",
    "typescript": "^5.8.2"
  },
  "scripts": {
    "build": "ray build",
    "dev": "ray develop",
    "fix-lint": "ray lint --fix",
    "lint": "ray lint",
    "prepublishOnly": "echo \"\\n\\nIt seems like you are trying to publish the Raycast extension to npm.\\n\\nIf you did intend to publish it to npm, remove the \\`prepublishOnly\\` script and rerun \\`npm publish\\` again.\\nIf you wanted to publish it to the Raycast Store instead, use \\`npm run publish\\` instead.\\n\\n\" && exit 1",
    "publish": "npx @raycast/api@latest publish"
  }
}

2. ESLint Config Issue

File: extensions/supermemory/eslint.config.js

The ESLint config syntax is incorrect for the current version:

const { default: raycastConfig } = require("@raycast/eslint-config")

module.exports = raycastConfig

3. API Error Handling Enhancement

File: extensions/supermemory/src/api.ts (lines 119-136)

The error handling could be more robust:

      throw new SupermemoryAPIError(errorMessage, response.status);
    }

    if (!response.headers.get("content-type")?.includes("application/json")) {
      throw new SupermemoryAPIError("Invalid response format from API");
    }

    const data = (await response.json()) as T;
    return data;

4. Memory Interface Inconsistency

File: extensions/supermemory/src/api.ts (lines 19-26)

The Memory interface doesn't align with the actual API response structure used in search:

export interface Memory {
  id: string;
  content: string;
  title?: string;
  url?: string;
  containerTag?: string;
  createdAt: string;
  chunks?: unknown[];
  metadata?: Record<string, unknown>;
}

Minor Suggestions

1. Add validation for URL format in the add-memory form
2. Consider adding a loading indicator during API key validation
3. Add keyboard shortcuts for common actions in the search view
4. Consider caching projects to reduce API calls

Security & Best Practices

✅ API key is properly secured in preferences
✅ Input validation is present
✅ Error messages don't expose sensitive information
✅ Proper TypeScript usage throughout

Verdict

This is a solid extension with good architecture and user experience. The main issues are configuration-related (ESLint config and dependency versions) rather than functional problems. Once these are fixed, it should work well for users.

Recommendation: Approve after addressing the configuration issues in package.json and eslint.config.js.

[MaheshtheDev]

@greptileai can you do a fresh review?

[greptile-apps]

I'll do a fresh review of this supermemory extension PR. Let me examine the code and structure.

Overall Assessment

This is a well-structured extension that integrates Supermemory's knowledge management system with Raycast. The code quality is good with proper TypeScript usage, error handling, and follows Raycast conventions.

Strengths

✅ Clean Architecture: Well-organized with separate API layer, proper TypeScript interfaces, and good separation of concerns

✅ Error Handling: Comprehensive error handling with custom error classes (SupermemoryAPIError, AuthenticationError) and user-friendly toast messages

✅ User Experience:

• Debounced search (500ms) for better performance
• Loading states and empty views
• Proper keyboard shortcuts
• Intuitive form validation

✅ Security: API key stored securely in preferences with proper validation

✅ Code Quality: Good TypeScript practices, proper imports, and consistent formatting

Issues Found

1. Minor: Incomplete changelog date format

The changelog shows {2025-09-27} which should be 2025-09-27 (without braces).

# supermemory-raycast Changelog

## [Initial Version] - 2025-09-27

- Added Supermemory integration with Add Memory and Search Memories commands
- Added project organization support for memories

2. Code Quality: Unused type definitions

Some interfaces in api.ts like Memory appear to be defined but not actively used in the current implementation.

3. Enhancement Opportunity: Error boundary

While error handling is good, the search component could benefit from better handling of network timeouts or API rate limits.

Suggestions for Improvement

1. Add retry logic for failed API requests
2. Consider adding memory caching for projects to reduce API calls
3. Add input validation for URL format in add memory form (though optional)
4. Consider adding batch operations for multiple memory management

Security & Privacy

✅ API key handling is secure
✅ No sensitive data logged
✅ Proper HTTPS usage

Documentation

✅ Good README with clear setup instructions
✅ Proper package.json metadata
✅ Screenshots included in PR

Verdict

This is a solid extension that's ready for the store with minor fixes. The code demonstrates good engineering practices and provides real value to users who want to integrate their knowledge management workflow with Raycast.

Recommendation: Approve with minor changelog fix.

[github-actions]

Published to the Raycast Store:
https://raycast.com/supermemory/supermemory

[raycastbot]

🎉 🎉 🎉

We've rewarded your Raycast account with some credits. You will soon be able to exchange them for some swag.