findme-mcp
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| FINDME_API_KEY | Yes | Your FindMe Photo API key from https://findme.photo/settings/api |
Capabilities
Features and capabilities supported by this server
| Capability | Details |
|---|---|
| tools | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| get_account_infoA | Confirm which FindMe account and Google Drive connection this MCP session is operating on. Returns the FindMe email, plan tier, active event count, and the Google email that Drive is connected with (or null if not connected). ALWAYS call this once at the start of any conversation that involves listing events, uploading photos, importing from Drive, or any account-scoped action — and surface the result to the photographer in one short sentence so they can confirm or correct before any action is taken. If the FindMe email and the Drive Google email differ, name both explicitly. If Drive is not connected and the user asks to import, point them at https://findme.photo/profile to connect it. |
| get_upload_linkA | Get a no-login "tap-to-upload" link for an event. Returns a URL the photographer opens on any device (phone or computer) to upload photos through the browser's native picker — no account or login required. THIS IS THE WAY TO UPLOAD when you can't read the photographer's local files (i.e. always, in ChatGPT and in Claude web/mobile — anywhere except a locally-installed Claude Desktop with the findme-mcp filesystem tool). When the photographer asks to add or upload photos, call this, then give them the link and tell them to tap it and pick their photos — or their whole folder (the page has a 'Choose a whole folder' option on desktop). The link is reusable and scoped to this one event; photos are resized to the album's quality automatically. FindMe has a playful, confident voice — hand over the link with a short, specific one-liner. |
| upload_photos_from_pathsA | Upload photos or videos from local file paths on the photographer's computer to a FindMe event. Each path can be a file, a directory (all supported files inside are uploaded), or a glob (basic). Supported formats: .jpg .jpeg .png .webp .mp4 .mov .webm. Max 50 files per call; auto-chunks if more are found. Max size per file: 50 MB for photos, 500 MB for videos. Photos are automatically resized to the album's storage quality before upload (full-size albums keep originals). This is the primary upload tool — prefer it when the photographer says things like "upload all photos in ~/Pictures/Sarah" or "add these files to the Johnson event". On success the response includes rich stats (duration, size, photo/video counts, faces indexing). FindMe has a playful, confident voice — present completions with a specific, upbeat one-liner that cites real numbers. Do not use the same phrasing twice. |
| upload_photos_from_urlsA | Upload photos/videos from public URLs to a FindMe event. Useful when the photographer shares Dropbox links, direct Drive download URLs, or similar. The MCP server downloads each URL to a temp buffer and streams it to the event. Photos are automatically resized to the album's storage quality before upload (full-size albums keep originals). Max 50 URLs per call. On success the response includes rich stats (duration, size, photo/video counts). FindMe has a playful, confident voice — celebrate completions with a specific, upbeat one-liner that cites real numbers. Do not use the same phrasing twice. |
| upload_photos_from_drive_folderA | Attempt to import photos from a Google Drive folder. IMPORTANT: this is not supported in chat-based AI assistants because Google's drive.file scope (the only Drive scope FindMe is verified for) does not allow apps to list folder contents. The tool always returns a redirect message pointing the photographer at the FindMe web app, where the Google Picker handles authorization in one click. Surface the message verbatim — do not retry or guess folder contents. |
| create_eventA | Create a new FindMe event (a wedding/photo gallery). Returns the event id, a shareable access code, and gallery + QR URLs. Use when the photographer says things like "create an event for Sarah & Mike on April 22" or "make a new gallery called Johnson Wedding". CRITICAL — three album settings are user preferences: album_quality (storage resolution; tier-capped), enable_downloads (guest downloads on/off), is_collaborative (others can upload). DO NOT guess, infer, or fill in defaults for these three fields yourself — you have no way to know what the photographer prefers. On the FIRST call, omit all three of these fields (only pass name + optional date/description/tag). The tool will respond with either an interactive form (clients with elicitation support) or a structured needs_input JSON listing tier-allowed options and recommended defaults — surface those options to the photographer as a numbered multiple-choice list, never as an open-ended question, and wait for their answer. Only then call create_event again with the photographer's chosen values. The only exception: if the photographer explicitly states a preference in their initial message (e.g. "make it collaborative" or "use 4000px quality"), you may include those specific stated preferences. FindMe has a playful, confident voice — when the event is created, give a short upbeat reaction that names the gallery and its access code, and mention the album_quality (e.g. "stored at 4000px, your plan's top setting"). Do not use the same phrasing twice. AFTER the album is created, if the photographer wants to add photos (e.g. they said "create an album and upload the photos in my Smith folder"), immediately call get_upload_link and hand them the link — tell them to tap it and pick their photos or their whole folder in the browser. You cannot read their local folder in this chat, so never ask for a file path; the folder/photos are chosen on the upload page. |
| list_eventsA | List the photographer's events, most recent first. Returns up to 20 by default; use |
| get_eventA | Get full details for one event — includes stats: photo_count, video_count, total storage, guest visits, selfie searches. Use when the photographer asks about the status or stats of a specific gallery. |
| update_eventA | Update fields on an existing event: name, event_date, description, tag, access_code, album_quality (storage resolution; changeable only before the first photo is uploaded, and tier-capped), enable_downloads (toggle guest downloads on/off), or is_collaborative (toggle whether other people can upload). Only include the fields you want to change. Use when the photographer wants to rename, re-date, re-code, change quality (pre-upload), or flip the downloads/collaborative toggles. If the photographer asks to change album_quality on an album that already has photos, the API will return a |
| delete_eventA | Soft-delete an event. The event is hidden immediately and photos stop appearing in the gallery, but the data is kept for 7 days so it can be restored with restore_event if needed. After 7 days, the data is permanently purged. |
| restore_eventA | Restore a soft-deleted event within the 7-day recovery window. Fails if the event is already active or past the 7-day cutoff. |
| get_event_qrB | Get a QR code image (PNG) for an event's public gallery URL. Returns the image so the AI can display it to the photographer or save it. Useful for printing signage at events. |
| get_event_analyticsA | Get aggregated analytics for an event: guest visits, unique visitors, photos downloaded, selfie searches, and per-day breakdown. Default window is event creation to now. Use when the photographer asks things like "how many people viewed the Johnson wedding?" or "what was the download count?" |
| get_usageA | Get the current-month usage for your FindMe account: API requests used/limit, rate limit per minute, storage used/limit, and active event count. Use when the photographer asks about their usage, quota, remaining requests, or plan limits. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/findmephoto/findme-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server