SFX Desk / API
Get a token

Driving SFX Desk from code

Everything the web app does is available over HTTP. The base URL is https://api.skillsafe.ai/v1/app-api, every request carries Authorization: Bearer <token>, and every response is the same envelope.

The task field comes first

This app has five lanes behind one endpoint. Every run body must carry a task field naming the lane - it is what the system prompt routes on. Send the wrong one and you get a valid package of the wrong kind; omit it and the model picks the closest lane and tells you which it chose.

One more shape trap: the run body is the input object. Do not wrap it in an {"input": ...} envelope - that returns 200 while hiding task from the model, which is the most confusing way this API can fail.

taskLaneFieldsSections returned
planTurn a list of picture events into a cue sheetbrief, knownSummary, The Cue Sheet, The Numbers, Reasoning, Next Step
placeWhere every file has to start, and where every hit landssheet, worrySummary, Verdict, Findings, Corrected Sheet, Next Step
nestWhat from means where, once Sequences nestsheetSummary, The Tree, Every Cue Resolved, What Moving Costs, Next Step
tailWhat the attack and the decay lose to the windowsheetSummary, The Attack, The Decay, What The Window Costs, Next Step
deliverDecide what changes: the times, the runway, or the ratesheet, fixedSummary, A Frame Move Fixes, Only Runway Or Rate Fixes, Nothing Fixes, Next Step

Only task and the lane's own required fields are mandatory: sheet on place, nest, tail and deliver; brief on plan. Every field is a string - there are no number fields on this app. sheet is the cue sheet itself: a header of KEY: value lines and up to three blocks, FILES:, SEQUENCES: and CUES:.

preroll= on a FILES line is the number that matters most. It is the distance in milliseconds from the start of the file to its transient, and it is the one number the engine cannot infer - a whoosh carries about 180 ms and a click about 12. Without it a file is placed as if it began at its impact, which is the exact mistake this app exists to catch, so an omitted pre-roll is reported on every answer rather than assumed quietly.

The header takes JOB, FPS and COMPOSITION. Without them the engine assumes 30 fps and 300 frames and says so, and every frame number in the answer is a function of the rate. COMPOSITION is the root durationInFrames, which is the outermost window and therefore what cuts a tail when no Sequence does.

A FILES line takes preroll=, dur=, peak= and kind=hard|soft, all in milliseconds except the peak in dBFS. A SEQUENCES line takes from= and dur= in frames plus in= naming its parent - and because useCurrentFrame() is relative, nested from values ADD, so the tree is what turns a from into an absolute frame. A CUES line takes file= (defaulting to the cue's own name), hit=, in= (defaulting to the root composition) and kind=.

A hit time may be written 1.225s, 1225ms, frame 96, f96, 96f, 00:00:01.225 or 01:23.4. A bare number with a decimal point is seconds and a bare integer is a FRAME, because hit=96 in a Remotion sheet means frame 96 - getting that wrong is wrong by a factor of the frame rate, so the form that was used is stated back on every cue. A hit quoted in frames has no rounding of its own, which halves the worst-case landing error for free.

Anything the reader cannot place is listed as a problem rather than skipped. A sheet whose cue lines were quietly swallowed would read as an empty composition, and every figure in the answer would be a confident statement about nothing.

A FRAME is reported as a bare integer, a time in milliseconds to one place, a landing error signed, a share as a percentage, a level in dBFS to two places, and a from value as from={N}. The engine works in frames and milliseconds: it has decoded no audio and looked at no waveform.

Add $model to any body to choose the model for that run: gpt-5.6-luna, gpt-5.6-terra (the default) or gpt-5.6-sol. Luna caps output at 4,096 tokens and will fail the place, nest and tail lanes rather than shorten them - a findings table, a corrected sheet, or a row per cue and per Sequence, is several thousand characters before the reasoning starts.

The response envelope

Success and failure have the same outer shape, so one check covers both.

{
  "ok": true,
  "data": {
    "...": "the result"
  }
}
{
  "ok": false,
  "error": {
    "code": "VALIDATION_ERROR",
    "message": "seconds should be number, got string",
    "details": {}
  }
}
HTTPerror.codeWhat it means
400VALIDATION_ERRORThe body was not a JSON object, or a declared field had the wrong type. A number field sent as a string is the usual cause.
401UNAUTHORIZEDNo token, or a token that has expired or been revoked. Mint a new one.
402INSUFFICIENT_CREDITSThe balance is below the run's minimum. Call /estimate first and compare hold_credits against /me.
404NOT_FOUNDWrong path, or a job id that does not belong to this token.
409CONFLICTAn Idempotency-Key replay whose body differs from the original request.
429RATE_LIMITEDToo many requests. Back off; do not tight-loop.
503UPSTREAM_UNAVAILABLEThe model provider is unavailable. Retry with backoff.

1. Get a token

Open /tokens.html in a browser and copy the token this app already holds - no developer console needed. A guest token is minted automatically and is enough for /me and /estimate; writing a package is metered and needs a personal token, which comes from signing in on that page.

Keep it in an environment variable rather than in source:

export SKILLSAFE_TOKEN="YOUR_TOKEN"

2. Check the session and the balance

GET /me is free. It returns only three fields: subject_type, subject_id and credits. Signed-in means subject_type == "user" - there is no username or email to test.

curl -sS -X GET "https://api.skillsafe.ai/v1/app-api/me" \
  -H "Authorization: Bearer $SKILLSAFE_TOKEN"

3. Price the run before making it

POST /estimate costs nothing, creates no job, and returns the worst-case cost. Compare hold_credits against the balance from step 2 before you submit: a 402 after the fact is avoidable. hold_credits is a reservation priced at the full output cap - the actual charge is usually far lower.

It also echoes model, model_alias and markup_bps, which is the authoritative check that a run is bound to the model you think it is. Estimate each lane separately: their prompts and caps differ, so their holds do.

curl -sS -X POST "https://api.skillsafe.ai/v1/app-api/estimate" \
  -H "Authorization: Bearer $SKILLSAFE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
  "task": "place",
  "sheet": "<FPS, COMPOSITION, FILES with preroll=, SEQUENCES, CUES; the grammar is in /llms.txt>",
  "worry": "it graded fine last time and this one will not come clean",
  "rules": "<the working rules for this lane, sent by the app>"
}'

4. Write a package

POST /run submits the job. Always send an Idempotency-Key: a network blip that replays the same request must not bill twice. A replay with the same key returns the stored result and is not charged again; a replay with the same key but a different body is a 409.

The response carries output.output (the Markdown package), charged_credits and truncated. If truncated is true the balance sat between min_credits and hold_credits and the output was cut short - render what arrived and say so rather than presenting it as complete.

curl -sS -X POST "https://api.skillsafe.ai/v1/app-api/run" \
  -H "Authorization: Bearer $SKILLSAFE_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
  "task": "place",
  "sheet": "<FPS, COMPOSITION, FILES with preroll=, SEQUENCES, CUES; the grammar is in /llms.txt>",
  "worry": "it graded fine last time and this one will not come clean",
  "rules": "<the working rules for this lane, sent by the app>"
}'

5. Stream a run

POST /run-stream is the same call with a text/event-stream response. Worth knowing before you build on it: from a server or from cURL you get event: delta frames carrying the output token by token; from a browser you get event: tick heartbeats and then one event: done with the whole output. Handle both, and treat ticks as liveness rather than progress.

Frame types are job (the job id), delta ({"text": "..."}), tick ({"t": seconds}), done, and error. An idempotent replay returns plain JSON with no stream at all, so check the content type before you start reading frames.

curl -sS -N -X POST "https://api.skillsafe.ai/v1/app-api/run-stream" \
  -H "Authorization: Bearer $SKILLSAFE_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Accept: text/event-stream" \
  -H "Idempotency-Key: cbd-$(date +%s)" \
  -d '{
  "task": "place",
  "sheet": "<FPS, COMPOSITION, FILES with preroll=, SEQUENCES, CUES; the grammar is in /llms.txt>",
  "worry": "it graded fine last time and this one will not come clean",
  "rules": "<the working rules for this lane, sent by the app>"
}'

6. Read the result

output.output is Markdown in the envelope this app's system prompt guarantees: every section is a level-two heading spelled exactly as listed in the lane table above, in that order; tables are GitHub pipe tables with the declared columns; prompts are in fenced blocks opened with three backticks and the word text; checklists are - [x] lines.

So parsing is a split on /^## / - but do it fence-aware, because a prompt block can legitimately contain a line starting with ##. Count the sections you got against the ones the lane declares: a short list means the run was truncated, not that the contract changed.

def sections(md):
    out, name, buf, fence = {}, None, [], False
    for line in md.split("\n"):
        if line.lstrip().startswith("```"):
            fence = not fence
        if not fence and line.startswith("## "):
            if name:
                out[name] = "\n".join(buf).strip()
            name, buf = line[3:].strip(), []
            continue
        if name:
            buf.append(line)
    if name:
        out[name] = "\n".join(buf).strip()
    return out

The artifact most callers want is the fenced text block inside ## The Sheet or ## Corrected Sheet - that is a complete sheet in the grammar above, so it can be fed straight back into another lane with nothing carried alongside it. Every other section is prose and tables meant to be read.

Rate limits and good manners