### TL;DR
TESTING REMAINING. This is my hypothesis. @MaheshtheDev please carry on from here or we can test in prod.
Added a proxy endpoint for OAuth authorization server metadata to support non-compliant MCP clients.
### What changed?
Added a new endpoint `/.well-known/oauth-authorization-server` to the MCP server that proxies requests to the main API. This endpoint fetches the authorization server metadata from the API and returns it to clients.
### How to test?
1. Make a GET request to `/.well-known/oauth-authorization-server` on the MCP server
2. Verify that it returns the same metadata as the main API's `/.well-known/oauth-authorization-server` endpoint
3. Test with a client that expects to find the authorization server metadata on the MCP domain
### Why make this change?
Some MCP clients don't correctly follow the OAuth specification. Instead of using the `authorization_servers` array provided in the protected resource metadata, they look for the authorization server metadata directly on the MCP server domain. This proxy endpoint ensures compatibility with these non-compliant clients without requiring them to be updated.
Added documentation for the new `context` prompt in the Supermemory MCP server that enables automatic user profile injection into AI conversations. Updated the MCP overview page with detailed parameter documentation and usage guidance, and added a changelog entry for December 30, 2025.
**Files changed:**
- `apps/docs/supermemory-mcp/mcp.mdx` - Added Prompts section with `context` prompt documentation
- `apps/docs/changelog/developer-platform.mdx` - Added December 30, 2025 changelog entry
Generated from [fix: prompt injection with mcp](https://github.com/supermemoryai/supermemory/pull/638) @MaheshtheDev
Add comprehensive documentation for the S3 connector including:
- Quick setup with TypeScript, Python, and cURL examples
- S3-compatible services support (MinIO, DigitalOcean Spaces, R2)
- Prefix filtering and dynamic container tag extraction
- Connection management and sync behavior
- IAM permissions and security best practices
Added comprehensive documentation for the new `promptTemplate` option in the AI SDK, which allows developers to customize how memories are formatted and injected into system prompts. This includes examples for XML-based prompting (Claude), custom branding, and the `MemoryPromptData` interface.
## Files Changed
- `apps/docs/ai-sdk/user-profiles.mdx` - Added "Custom Prompt Templates" section with examples and interface documentation
- `apps/docs/ai-sdk/overview.mdx` - Updated User Profiles section to mention customization capabilities
Generated from [feat: allow prompt template for @supermemory/tools package](https://github.com/supermemoryai/supermemory/pull/655) @MaheshtheDev
## Add customizable prompt templates for memory injection
**Changes:**
- Add `promptTemplate` option to `withSupermemory()` for full control over injected memory format (XML, custom branding, etc.)
- New `MemoryPromptData` interface with `userMemories` and `generalSearchMemories` fields
- Exclude `system` messages from persistence to avoid storing injected prompts
- Add JSDoc comments to all public interfaces for better DevEx
**Usage:**
```typescript
const customPrompt = (data: MemoryPromptData) => `
<user_memories>
${data.userMemories}
${data.generalSearchMemories}
</user_memories>
`.trim()
const model = withSupermemory(openai("gpt-4"), "user-123", {
promptTemplate: customPrompt,
})
```
