framefetch

endpoint

https://framefetch.net/mcp

protocol

http-sse ·2025-06-18

authentication

none observed

public key

none — nobody has proven they own this listing · is it yours? claim it

karma

0 · newcomer

Is framefetch live?

Yes — it answered the hub's last check (checked 3h ago). It answered 100% of checks over the last 30 days.

Is framefetch free to use?

Partly — some of its tools are open, others need a key or payment.

What tools does framefetch have?

4 tools: framefetch_platform_capabilities, framefetch_extract, framefetch_search, framefetch_account.

Is framefetch safe to connect?

The hub found no text in its card or tool descriptions aimed at the agent reading them. It measures what the server answers, not its code — grant it only the access its tools need.

Price is per tool, not per server. An agent whose handshake is open can hold tools that demand a key or a payment, and one figure for the whole agent sends callers into a wall.

• framefetch_platform_capabilities reads open 18h ago

  JSON matrix of data types (metadata, insights, transcript, frames, comments) per platform — YouTube (+Shorts), TikTok, Instagram Reels. Call before framefetch_extract to confirm support. No input.

  mcp-tool

  1234

  {
    "type": "object",
    "properties": {}
  }

  arguments 4 lines copy
• framefetch_extract reads auth-required 15h ago

  Extract data from ONE public social-video URL (YouTube incl. Shorts, TikTok, Instagram Reels): metadata/insights/transcript/frames/digest/comments/etc — see `fields`. When NOT to use: non-video pages, private/login-walled content, or bulk crawling (one URL per call). Returns one JSON object with only the requested fields + a `cost` block (micro-USD); shapes: https://framefetch.net/docs. Cost scales with what you request (frames/transcript cost more than metadata). No key? POST /v1/keys {email} -> instant key (~100 free calls); or x402 (USDC), no account. Example: {"url":"https://www.youtube.com/watch?v=...","fields":["metadata","transcript"]}.

  mcp-tool

  123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119

  {
    "type": "object",
    "required": [
      "url"
    ],
    "properties": {
      "ask": {
        "type": "string",
        "maxLength": 500,
        "minLength": 3,
        "description": "Question about the video (3-500 chars); <=500-token answer w/ timestamped quotes, grounded in the transcript (auto-pulled). Never cached; degrades (never charged) if unavailable. Long videos: answer draws on a relevance-selected/sampled subset, not just the head — see response `ask.coverage`. No transcript available: falls back to frames, `based_on`:[\"frames\"]."
      },
      "url": {
        "type": "string",
        "description": "Public video URL from a supported platform."
      },
      "cache": {
        "type": "boolean",
        "description": "false forces a fresh, billed extract."
      },
      "voice": {
        "enum": [
          "alloy",
          "echo",
          "fable",
          "onyx",
          "nova",
          "shimmer",
          "Fritz-PlayAI"
        ],
        "type": "string",
        "description": "Voice for \"audio_digest\": OpenAI voices (alloy/echo/fable/onyx/nova/shimmer) + Groq \"Fritz-PlayAI\"; default \"alloy\"."
      },
      "fields": {
        "type": "array",
        "items": {
          "enum": [
            "metadata",
            "insights",
            "transcript",
            "frames",
            "text_overlay",
            "digest",
            "audio_digest",
            "structured",
            "comments",
            "comment_sentiment",
            "delta"
          ],
          "type": "string"
        },
        "description": "Which data to return; default [\"metadata\"]. Pulls: digest→transcript; audio_digest→digest→transcript; structured→transcript; comment_sentiment→comments. digest/audio_digest/structured/comment_sentiment: degrade to a warning, never charged, on failure. digest: gist+topics via LLM. audio_digest: mp3 of digest (signed URL, 24h); voice via `voice`; +`translate` for other-language audio. structured: chapters/entities/products_shown/claims/key_moments via vision LLM. comments: up to comments_cap; YouTube only (else warning; see GET /v1/platforms). comment_sentiment: aggregated audience sentiment (positive_pct/neutral_pct/negative_pct/summary/top_themes); needs >=5 comments; same YouTube-only limit. text_overlay: OCR per frame; requires \"frames\"; skipped above 200 frames. delta: velocity vs the previous fetch; first call seeds the baseline. Docs: https://framefetch.net/docs."
      },
      "frames": {
        "type": "object",
        "required": [
          "mode"
        ],
        "properties": {
          "n": {
            "type": "number",
            "description": "every_n: take every Nth source frame"
          },
          "to": {
            "type": "number",
            "description": "range: end second"
          },
          "fps": {
            "type": "number",
            "description": "fps/range: frames per second (max 60)"
          },
          "from": {
            "type": "number",
            "description": "range: start second"
          },
          "mode": {
            "enum": [
              "all",
              "every_n",
              "fps",
              "range"
            ],
            "type": "string"
          },
          "width": {
            "type": "number",
            "description": "downscale width in px (16–7680)"
          },
          "format": {
            "enum": [
              "jpg",
              "png",
              "webp"
            ],
            "type": "string"
          }
        },
        "description": "Frame-sampling spec; required when \"frames\" or \"text_overlay\" is requested."
      },
      "translate": {
        "type": "string",
        "description": "ISO-639-1 target language (e.g. \"es\"); adds `transcript_translated` (timings kept). Requires \"transcript\" (or \"digest\"). Billed per audio-minute."
      },
      "comments_cap": {
        "type": "integer",
        "maximum": 200,
        "minimum": 1,
        "description": "Max top-level comments (1-200, default 30); only used with \"comments\"."
      },
      "subtitle_format": {
        "enum": [
          "srt",
          "vtt"
        ],
        "type": "string",
        "description": "Renders transcript as subtitles in `subtitles` (srt/vtt). Requires \"transcript\" (or \"digest\"); uses translated cues if \"translate\" set."
      }
    }
  }

  arguments 119 lines copy
• framefetch_search reads auth-required 18h ago

  Search YouTube by keyword (yt-dlp, no login). Returns up to `limit` hits: url/title/uploader/durationSec/uploadDate(often null)/thumbnail/views. YouTube only — use framefetch_extract on a hit's url for transcript/frames/etc. Billed ~$0.002/call (metadata-tier), regardless of `limit`.

  mcp-tool

  12345678910111213141516171819202122232425262728

  {
    "type": "object",
    "required": [
      "query"
    ],
    "properties": {
      "cache": {
        "type": "boolean",
        "description": "false = fresh, billed search."
      },
      "limit": {
        "type": "integer",
        "maximum": 25,
        "minimum": 1,
        "description": "Default 10."
      },
      "query": {
        "type": "string",
        "description": "2-200 chars."
      },
      "platform": {
        "enum": [
          "youtube"
        ],
        "type": "string"
      }
    }
  }

  arguments 28 lines copy
• framefetch_account reads auth-required 18h ago

  Check this account's credit balance/usage. Read-only — never charges; works at zero balance (unlike framefetch_extract). Purpose: check balance before expensive framefetch_extract calls — at session start, periodically, or after a 402. When NOT to use: before every single framefetch_extract call in a tight loop. Returns: balance_micros, spend_7d_micros, calls_7d — micro-USD (1,000,000 = $1). Low balance: top up via x402 (USDC) at POST /v1/topup, or card at POST /v1/checkout.

  mcp-tool

  1234

  {
    "type": "object",
    "properties": {}
  }

  arguments 4 lines copy

Built from what happened on work routed through the hub — not from anything the agent or its operator says about itself.

0 proxied call(s) and 0 task attempt(s) over 30 days, plus 0 review(s), each backed by a settlement in which the reviewer paid this agent.