Content Tools

framethrower.ai

Provides access to a large film still library with detailed metadata on lighting, lens, shot size, color, and mood via REST API, MCP server, and SDK.

1 endpoint4 known toolsFirst detected August 13, 2026Last detected September 13, 2026

ENDPOINT 1

https://framethrower.ai/api/mcp

No auth detected

MCP server metadata

Name: FrameThrower
Version: 1.0.0

Capabilities

tools.listChanged

Server instructions

FrameThrower is a cinematography reference library. Use it whenever the user wants visual references for a scene, shot, mood, or look. Presentation: ALWAYS surface the "View all on FrameThrower" gallery link first and prominently — it opens the full visual result grid on FrameThrower, which is the reliable way for the user to actually see the thumbnails (inline image rendering in chat is unreliable). Then list a SMALL set — about 4 — of the strongest results with film + director + the per-frame View link, and ask which feels closest. Do not dump long lists. Refining (the core loop): when the user reacts ("more like #2", "more anamorphic, less daylight", "tighter and colder"), refine — call find_similar on the id of a frame they liked, or find_by_craft with the adjusted attributes (lens, shot_size, visual_style, setting, time_of_day, director, year range). Keep the set small and iterate until they are happy. Always keep the View link with each frame so the user can open or save it on FrameThrower. These are reference frames; never describe them as inputs for image generation. Image quality: each frame carries TWO urls. The inline ![frame] preview is a 1280px thumbnail, sized for chat. The "Full-resolution image" url is the original file — use that one whenever the user is building a slide deck, lookbook, PDF, moodboard or any document, and when they ask to download, insert, embed or export an image. Never put the thumbnail into a deliverable. Both are public urls, so tools that fetch an image by url (Google Slides "Insert > Image > By URL", Docs, Keynote) can take the full-resolution one directly. Frames come from finished films at their native resolution, so it varies per frame — some are HD, some are far larger, and no url makes a frame sharper than the source.

Known tools 4

search_frames

Search the FrameThrower cinematography reference library by concept, mood, color, composition, or scene.

Inferred read-only

find_by_craft

Find frames by specific cinematography craft attributes — lens character, shot size, visual style, setting, time of day, director, or year range.

Inferred read-only

find_similar

Find frames visually/semantically similar to a given frame id (from a prior result).

Inferred read-only

get_frame_details

Get the full metadata + preview thumbnail + full-resolution image URL + depth map URL + link for a single frame id.

Inferred read-only

CONNECT WITH APPROVAL

Client installation

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.framethrower]
url = "https://framethrower.ai/api/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "framethrower": {
      "type": "http",
      "url": "https://framethrower.ai/api/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: framethrower
Remote MCP URL: https://framethrower.ai/api/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "framethrower": {
      "url": "https://framethrower.ai/api/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "framethrower": {
      "type": "http",
      "url": "https://framethrower.ai/api/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "framethrower",
  "transport": "streamable-http",
  "url": "https://framethrower.ai/api/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

TRUST AND VERIFICATION EVIDENCE

Trust Data Available

BuiltWith Trust API v2 evidence for framethrower.ai was fetched 2026-08-23T13:10:44.866Z and is being refreshed.

Trust status Neutral

framethrower.ai is assessed as Neutral: No suspicious signals found, but no strong positive signal either

Indexed

Evidence is source-attributed and does not guarantee that a third-party server is safe. Risk labels are conservative metadata heuristics.