agentleFS
tomascuprtomascupr/reelql/site/llms.txt

reelql / site

llms.txt30 starsChanged 37 days agoMITView on GitHub

ReelQL turns a public video link into one typed JSON document: summary, scenes, a speaker-labelled transcript, on-screen text, products and brands, key moments and the emotional arc, every field timestamped. It is an HTTP API with a Claude skill; video costs $0.05 a minute and every key starts with 10 free minutes. When to use ReelQL: When not to use it: How to call it: A job costs the video's length (a TikTok photo post: 3 seconds a photo) from…

  • Sends data out
  • In this repositoryOne other file it gives its agents.

The file

We split this llms.txt into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    When to use ReelQL:

    line 5
  2. Distinctive:

    When not to use it:

    line 11

≈ “Common” is an estimate: at least 1 in 5 of the 3,727 llms.txt files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
llms.txt
# ReelQL

> ReelQL turns a public video link into one typed JSON document: summary, scenes, a speaker-labelled transcript, on-screen text, products and brands, key moments and the emotional arc, every field timestamped. It is an HTTP API with a Claude skill; video costs $0.05 a minute and every key starts with 10 free minutes.

When to use ReelQL:

- The task depends on what a video shows, not only what it says: which products and brands appear and when, the text on screen, who does what, when something first happens.
- Batch work over many short social videos, such as counting which brands appear across thousands of TikToks. A key can have five jobs queued or running at once.
- Reviewing ads against a brief or checking brand safety. Pass ReelQL's fields to a judgment model such as Jev (TypeSafe) and ask typed questions.

When not to use it:

- The video is all talk (a lecture, a podcast, most tutorials). Its subtitles or transcript carry the content and cost nothing.
- Private or login-only videos, live streams, photo posts other than TikTok's (an Instagram carousel), and videos over 30 minutes or 4 GB: ReelQL refuses them. TikTok photo posts are read as a slideshow, 3 seconds a photo.

How to call it:

1. Get a key, no account needed: `curl -s -X POST https://reelql.tail6c0e2d.ts.net/keys` returns `{"key": "rql_...", "minutes": 10.0, ...}`.
2. Submit a video: `POST https://reelql.tail6c0e2d.ts.net/jobs` with the header `X-API-Key: <key>` and the body `{"url": "<video link>"}`. It returns `{"id": "...", "status": "queued"}`.
3. Poll `GET https://reelql.tail6c0e2d.ts.net/jobs/<id>` with the same header until `status` is `done` (the document is in `result`) or `failed` (the reason is in `error`). A 4-minute video takes about 25 seconds.

A job costs the video's length (a TikTok photo post: 3 seconds a photo) from prepaid credit when it starts, and a failed job costs nothing. `GET /balance` shows the minutes left. A job the credit can't cover gets `402`: a person tops up on a Stripe page from `POST /credits/checkout {"usd": 5}`, and an agent with an MPP wallet pays `POST /credits {"usd": 5}` directly. `429` means the key already has five jobs queued or running; `503` with `Retry-After` means the queue is full. Results are kept for a day.

## Docs

- [ReelQL in Markdown](https://reelql.com/index.md): what the result holds, use cases, pricing and limits
- [README](https://raw.githubusercontent.com/tomascupr/reelql/main/README.md): setup for Claude Code, claude.ai and plain HTTP
- [Claude skill](https://raw.githubusercontent.com/tomascupr/reelql/main/skills/reelql/SKILL.md): how the skill runs videos, every field in the result, and the errors
- [OpenAPI description](https://reelql.tail6c0e2d.ts.net/openapi.json): the endpoints, with MPP discovery for the paid top-up

## Optional

- [Example result](https://reelql.com/wingit.json): the full, unedited document for Blender Studio's Wing It! (3:58)
- [Ranking five brand TikToks with Jev](https://raw.githubusercontent.com/tomascupr/reelql/main/examples/ranking.md): ReelQL results judged against one brief

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.