AI agents (MCP)
Crumbcast is an MCP server, so AI agents and coding assistants (Claude Code, Cursor, your own agent) can work with a project's videos and live streams: find a video, import one, caption it, cut it, read its numbers, go live.
The server is at https://your-crumbcast/api/v1/mcp (Streamable HTTP). It signs in with a
project's API token, sent as the bearer token, like the API's requests,
and it's held to the same rules: the token's project only, at the token's role. Give an
agent a viewer token to let it look without changing anything.
Connecting
Claude Code:
claude mcp add --transport http crumbcast https://your-crumbcast/api/v1/mcp \
--header "Authorization: Bearer cc_…"
Clients configured with JSON (Cursor, Claude Desktop through mcp-remote, and others):
{
"mcpServers": {
"crumbcast": {
"url": "https://your-crumbcast/api/v1/mcp",
"headers": { "Authorization": "Bearer cc_…" }
}
}
}
Clients that only sign in with OAuth, such as claude.ai's connectors, can't connect yet.
Tools
| Tool | Does | Role |
|---|---|---|
project_overview | The project at a glance: videos, storage, views, live streams | viewer |
list_videos | The project's videos, by title words, folder (or none), metadata, still processing or failed, a page at a time | viewer |
get_video | A video's formats, captions, chapters, links, and what may be done to it now | viewer |
import_video | A new video from a public URL, with a workflow and metadata if you like | editor |
update_video | Its title, description, folder, labels, playback policy and password, chapters, metadata, ad tag | editor |
delete_video | Deletes a video: into the project's trash if it keeps one, else for good | editor |
list_trash, restore_video | The deleted videos the trash keeps, and bringing one back as it was | viewer, editor |
move_videos, label_videos | Moves videos into a folder, or adds a label to them or takes it off | editor |
create_folder, create_label | A new folder or label to sort videos into | editor |
generate_captions | Captions from the video's speech | editor |
get_transcript | What's said in a video, from its captions | viewer |
update_captions | A caption track's cues, corrected | editor |
suggest_chapters | Chapters read from its captions, to look over before saving | editor |
clip_video | A new video of the parts kept | editor |
trim_video | Cuts the video itself to the parts kept | editor |
retry_encoding | Encodes a failed format's failed parts again | editor |
set_thumbnail | A frame of the video as its thumbnail | editor |
run_workflow | Runs one of the project's workflows on a video already in | editor |
list_folders, list_labels, list_workflows, list_ad_tags | What a video can be sorted into, have done to it, or play | viewer |
list_playlists, create_playlist, update_playlist | Playlists, and the videos in them in play order | viewer, editor |
share_link | A watch link with a viewer link's token in it, for a signed or password video, live stream or playlist | editor |
get_analytics | Views and watch time, of the project or one video | viewer |
list_live_streams, get_live_stream | Live streams, whether they're live, and where to push to | viewer |
create_live_stream, update_live_stream | A live stream and its RTMP, SRT and WHIP ingest; its title, recording, latency and policy | editor |
delete_live_stream | Deletes a live stream; its recordings stay | editor |
Each tool describes its answer (its output schema), so an agent knows what it gets back
before it calls. list_videos and list_trash answer a page at a time: has_more says
whether there's another, and next_cursor, passed as after with the same filters and
order, asks for it.
Processing runs in the background, as it does through the API: a tool that starts it
answers at once, and the agent checks on it with get_video. What the API refuses, a
tool does too, with the same reason (403: …, 402: …), for the agent to read.