01
Save the question
Create a monitor, classifier, scorer, summarizer, extractor, or timestamped-moment scanner. Edits create traceable versions.
POST /v1/scanners Recorded-media intelligence for builders
Ask once. Scan every recording.
Save one question, run the same versioned scanner across authorized video or audio uploads, and turn long recordings into timestamped answers or editorially selected video moments.
Video + audio uploadsSelected time windowsSix scanner methodsExplicit rendering
"upload_id": "upl_...",
"metadata": {"cohort": "new-user"}
// GET /v1/runs/run_.../observations
{
"semantics": "timestamped_moment",
"value_type": "object",
"value": {"moment_id": "mom_...", "event_type": "setup_blocked"},
"relevance_score": 0.91,
"title": "Repeated setup failure",
"evidence": {"start_ms": 84210, "end_ms": 109440}
}One API primitive for
Session recordingsAudio interviewsSupport callsGameplay VODsResearch archives
The useful layer after recording
Go from hours of footage to evidence your product can use.
Define a scanner once, run it against completed private uploads, and query observations across runs without losing source or version provenance.
02
Find the moments worth keeping
Use video find_moments for the same editorial moment-selection quality as ChatClipThat, across the full source or a selected time window.
GET /v1/observations03
Render only when requested
Scanner runs never publish media automatically. Video find_moments returns Moment IDs you can explicitly send to the render API.
POST /v1/rendersAPI pricing
Pay from the same credit wallet you already use.
API usage is metered from your ChatClipThat account wallet. A paid account is required to issue API credentials, and the console shows only the checkout choices currently available to your account.
1 creditper minute analyzed
Usage rounds up to the next minute. A video find_moments window bills only its selected duration; without a window, the full upload is measured and billed. Each scanner run is billed independently.
9:01 = 10credits for one scanner
Run three saved scanners against that same upload and the total is 30 credits. Reusing an upload avoids another transfer, not another analysis charge.
Includedno additional API credits
Polling and structured JSON do not add charges. Video find_moments results can be rendered explicitly; scanner completion itself never publishes media, and audio has no render path.
Current access and limits. Paid accounts receive one proof analysis for one window up to 2 hours and one proof render totaling up to 60 seconds. Paid analyses have no fixed duration ceiling: each request must name an explicit window, fit the available credit balance, and stay within the account source-size limit. Longer windows use the same bounded chunk pipeline automatically. Video is capped at 8 GiB for the proof tier and 10 GiB for paid tiers; audio is capped at 2 GiB. Beginner, Hobby, and Creator allow 1, 2, and 3 concurrent runs with source retention of 7, 30, and 90 days. Enterprise limits are contracted. ChatClipThat system failures are refunded automatically.
Built for recorded experience data
An evidence-search layer for recordings of many kinds.
Use the same saved-scanner primitive for authorized video and audio—not only browser sessions and not only highlight reels.
- Ask about visible evidence in video or spoken evidence in audio
- Keep every answer tied to a grounded timestamp range
- Distinguish explicit monitor false from unanswerable zero results
- Reuse one private upload for several independent questions
01DefineSaved, versioned question
02UploadPrivate video or audio
03ObserveTyped evidence answers
04ActSearch, review, or automate
Make recorded video queryable
Your next workflow starts with one saved question.
Create an account, issue a revocable server-side key, privately upload an authorized recording, and run your first scanner. Explore concrete patterns for product research, interviews, gameplay, support, and media archives.