Video understanding for AI agents
Agents can read, but they can't watch. Give ReelQL a video link and get back one typed JSON document: scenes, speech, on-screen text, products and the emotional arc, each with its timestamp.
Every key starts with 10 free minutes, and you don't need an account to make one.
Now screening · a real run, unedited
Everything around the picture comes from ReelQL's JSON for this Blender Studio short. It watched the film in 22 seconds.
How it works
Paste a public link. ReelQL fetches the video, transcribes it with speaker labels, watches it 30 seconds at a time and writes one JSON document.
Jev, TypeSafe's judgment model, reads only text, so it judges a video through ReelQL's fields. Ask it typed questions, such as whether a video is on brief or brand-safe, and it answers with probabilities.
Sort, filter, route or flag on those numbers instead of parsing prose.
From link to JSON, on real runs.
What it's for
ReelQL reads what's on screen: products, logos, text and what people do. Speech is in the document too, but for a video that is all talk, a transcript does the job.
Brand scouting
A grocery buyer runs tens of thousands of TikToks through ReelQL to find brands worth stocking. Count the brands in each week's results: the names that keep climbing are the shortlist. A 30-second TikTok costs 2.5 cents, and a key runs five videos at a time.
brands products[].appearances on_screen_text
Ad review
Does the product show up early, and does the tone fit? ReelQL turns the video into fields and Jev answers each question with a probability, so you can sort a stack of ads instead of watching each one.
key_moments story.tone products transcript
Creator vetting
Before you sponsor a creator, run their recent posts. The transcript, the text on screen and what happens in each scene come back with timestamps, so you can check anything that looks off.
transcript on_screen_text scenes[].action
Skip it when
Lectures, podcasts and most tutorials live in their audio. Their subtitles are free and tell you as much, so start there. ReelQL is worth paying for when what matters is on screen.
The output
An excerpt from the Wing It! run above. The full document has twelve analysis fields, the speaker-labelled transcript, the video's metadata and per-step timing.
{"schema_version": "1.2","video": {"title": "WING IT! - Blender Open Movie", "channel": "Blender Studio", "duration_s": 238.0},"timing": {"total_s": 22.4},"analysis": {"summary": "An uptight cat engineer preparing a spacecraft for launch is interrupted by an enthusiastic brown dog who hijacks the controls, leading to a chaotic and comedic flight through the sky before crashing back into the barn they started in.","characters": [{"id": "c1", "name_or_label": "Grey Cat", "first_seen_s": 0},1{"id": "c2", "name_or_label": "Enthusiastic wannabe-pilot", "first_seen_s": 30}],"products": [{"name": "Spacesuit","brand": null,3"category": "Costume/Equipment","appearances": [{"t_s": 22.6, "how_shown": "Worn by Cat Engineer during exit scene.", "prominence": "medium"}2]}],"emotional_arc": [{"t_s": 87.0, "emotion": "excitement", "intensity": 5, "cue": "The dramatic launch of the rocket and the subsequent freefall."},4{"t_s": 102.0, "emotion": "fear", "intensity": 5, "cue": "Zero-gravity chaos"}],"on_screen_text": [{"t_s": 232.0, "text": "Licensed as Creative Commons Attribution 4.0 © Blender Foundation - studio.blender.org/wing-it"}5]},"speech": {"transcript": [{"start": 70.48, "end": 78.24, "speaker": "SPEAKER_00", "text": "I was looking for that."}]}}
brand and advertiser are null when nothing in the video supports them.Also in the document: story, audio, chapters, scenes, key_moments and timing. Every field is listed in the skill file.
With Jev
The skill's bundled script sends ReelQL's fields to Jev with five questions per video and sorts on the answers. With both keys set, ask Claude to rank some ads.
Family-friendly ad that shows the product in use
| # | Video · advertiser | On brief | Hook | Product | Payoff | Unsafe |
|---|---|---|---|---|---|---|
| 1 | samsungSamsung | 0.32 | 0.65 | 0.55 | 1% | |
| 2 | psgGoogle | 0.28 | 0.59 | 0.03 | 2% | |
| 3 | fifaworldcupLay's | 0.56 | 0.29 | 0.98 | 2% | |
| 4 | starbucksStarbucks | 0.83 | 0.09 | 0.12 | 2% | |
| 5 | natgeonatgeo | 0.49 | 0.01 | 0.01 | 1% |
Output of scripts/jev.py on the five TikToks in examples/ (jev-1.13.0, 2026-09-25). Jev's scores vary a little from run to run. Jev is made by TypeSafe; ReelQL is not affiliated with them.
from typesafe_sdk import Choice, Noul, Score, TypeSafeClient a = video["analysis"] # a ReelQL result brief = "Upbeat, family-friendly, shows the product in use" state = {"summary": a["summary"], "tone": a["story"]["tone"], "brands": a["brands"], "moments": {str(i): m["what"] for i, m in enumerate(a["key_moments"])}, "transcript": " ".join(s["text"] for s in video["speech"]["transcript"])} with TypeSafeClient() as jev: # reads TYPESAFE_API_KEY r = jev.system_one(state=state, questions={ "unsafe": Noul(instructions="Does `transcript` contain profanity, violence or adult content?"), "on_brief": Noul(instructions={"brief": brief, "question": "Does the video match `brief`?"}), "fit": Score(instructions={"brief": brief, "question": "How well does `tone` fit `brief`?"}, criteria=["Contradicts it", "Neutral", "Clearly fits"]), "best": Choice(instructions={"brief": brief, "question": "Which of `moments` best matches `brief`?"}, criteria={**{k: None for k in state["moments"]}, "none": None}), }) print(r.nouls["unsafe"].noul, r.nouls["on_brief"].noul, r.scores["fit"].score, r.choices["best"].choice)
| Recipe | ReelQL fields in | Jev question out |
|---|---|---|
| Brand safety | transcript, on_screen_text, scenes[].action | One yes/no per hazard, plus a severity score |
| On-brief check | summary, story.tone, story.themes, advertiser | "Matches the brief?" and a fit score |
| Cut-down picker | key_moments, scenes | Which moment best matches the brief |
| Rank a batch | The same fields for many videos | Hook, product clarity and payoff scores, then sort in code |
| Verify placements | products[].appearances, transcript | Is each product claim supported by the evidence? |
Send Jev only the fields a question needs: a long video's transcript can exceed its context. Do numeric comparisons, like views and likes, in code.
Get started
claude plugin marketplace add tomascupr/reelql claude plugin install reelql@reelql export REELQL_API_KEY=<your key> export TYPESAFE_API_KEY=<your Jev key> # optional: rankings and judgments
"Analyze this video <url>" gets you a short brief. You can also ask something specific, like "when does the product first appear?" Jev keys come from console.typesafe.ai.
On Team and Enterprise plans: download reelql.zip, upload it under Settings → Capabilities → Skills, and paste your key when Claude asks.
An organization owner first has to add reelql.tail6c0e2d.ts.net to the domains code execution may reach. On Pro and Max the sandbox reaches only package registries, so use Claude Code there.
API=https://reelql.tail6c0e2d.ts.net; H="X-API-Key: $REELQL_API_KEY" curl -s -H "$H" -H 'Content-Type: application/json' $API/jobs -d '{"url": "https://www.youtube.com/watch?v=..."}' # {"id": "6b0550c3dc9f", "status": "queued"} curl -s -H "$H" $API/jobs/6b0550c3dc9f # queued, running, then done (with "result") or failed (with "error")
Works from any agent that can make HTTP requests. A key without a browser: curl -s -X POST https://reelql.tail6c0e2d.ts.net/keys. GET /openapi.json describes the API, including the paid endpoint for MPP clients.
A key comes with 10 free minutes of video, without an account or a card.
Copy it now. ReelQL keeps only a hash of it, so a lost key can't be recovered.
For Claude Code or curl:
Pricing
You pay for the length of each video from prepaid credit. There is no subscription and no invoice.
GET /balance shows the minutes left.POST /credits speaks MPP, the Machine Payments Protocol: it answers 402 with a payment challenge, and an agent with an MPP wallet pays and retries.503 with Retry-After. A queued job's status shows its place in the queue.