List Mixpeek's namespace templates and their manifest URLs at https://mixpeek.com/templates/index.json (no key). To deploy one, fetch its manifest and call validate_manifest, then apply_manifest, on the Mixpeek MCP server (https://mcp.mixpeek.com/mcp). Every MCP tool needs an API key; an agent can get one with POST https://api.mixpeek.com/v1/signup/agent.7 templates
Stills are illustrative. The boxes are authored to show where each template's decisions fire.
sceneclearedMillions of scenes you can move through instead of search. Footage, ads and editor assets land in three clean collections, and scenes are clustered one partition at a time so every run stays under the sample-size caps.
Face indexedScene embeddedclearedSearch every scene of every ad you have run, and find the ones that keep earning spend
face 0.93 · est. 16 to 18 · borderlinereviewSegment a video corpus, extract a keyframe record for every segment, and match brand marks against a reference set you upload. Face and public-figure matching run on a dedicated or customer-hosted deployment. This deploy is the extraction layer a moderation policy sits on, and the page carries a state on every part so you can see which policy pieces are yours to add.
scene segmentstranscriptclearedAn article goes in, a ranked playlist of video comes out, and every match carries the entities, keywords and categories that produced it. Both sides are embedded twice, once by a text model and once by a multimodal one, so nothing is embedded at query time and a clip in another language still matches.
participantscreen state at that secondclearedModerated sessions become a corpus you can question. The transcript is joined to what was on screen at that second, so one query reaches every session at once and returns the moment rather than the recording.
Camera A · searchableCamera B · processingclearedTurn always-on RTSP cameras into a searchable monitoring wall where quiet footage is skipped before extraction, kept segments carry gate verdicts, and email alerts fire from retriever results.
overcast sky, read as nightbrake lights in daylightreviewFind the data your model is missing. Turn model failures into training and evaluation datasets with multimodal search and sampling.
Or let Mixpeek choose
Say what your files are and what you want from them. Mixpeek writes the namespace for you, with the extractors, models and retrievers, checks it against the catalog and prices it before you apply it.
To go further, the manifest loop tries other models, settings and retriever pipelines, scores each one on labelled examples it holds back, and hands you the best as a manifest you can deploy.