Kelvin Desk / API
Get a token

Driving Kelvin 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
planChoose the balance, then solve the gelsbrief, knownSummary, The Sheet, The Setting, Reasoning, Next Step
checkWhat every source lands at, and what the mix issheet, worrySummary, Verdict, Findings, Corrected Sheet, Next Step
mixWhat the sources really mix tosheetSummary, What Each Source Contributes, Mired Against Kelvin, The Spread, Next Step
gelsWhich gel each source actually needssheetSummary, Every Source, What Each Gel Is Worth Here, Where Nothing Fits, Next Step
deliverDecide what changes: the balance, the gels, or the fixturessheet, fixedSummary, Moving The Balance Fixes, Only A Gel Or A Fixture Fixes, Nothing Fixes, Next Step

Only task and the lane's own required fields are mandatory: sheet on check, mix, gels and deliver; brief on plan. Every field is a string - there are no number fields on this app. The sheet is two blocks, SETUP and SOURCES, and the grammar is in /llms.txt and in the free panel on the app itself.

A SOURCES row is id | what it is | kelvin | level | gel | why. A temperature accepts 5600, 5600K or 5.6k. The level is a weight, not a measurement - it decides how much each source counts toward the mix, so any consistent unit works and the absolute values cancel.

The gel column accepts a name (none, quarter cto, half ctb, full cto, double cto), a Lee number (Lee 204), or an explicit shift (+42 mired) for a fixture whose own temperature is being changed rather than filtered. Several sources can carry the same gel and each is computed from its own starting point, which is the whole point: a gel is a constant shift in mireds and therefore a variable one in Kelvin.

SETUP takes name, balance, target and tolerance. balance is what the camera is set to, and balance | mix is the shorthand for "whatever the camera would auto-balance to" - it makes the sheet report how far each source sits from the average rather than from a number somebody chose. tolerance is the residual cast you will accept, in mireds; it defaults to 5, which is about the smallest difference visible side by side on a white card. Leaving the balance out entirely is reported as a warning, because every cast on the page is measured from it.

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 check and split lanes rather than shorten them - a findings table, a corrected sheet and two ranked tables 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": "check",
  "sheet": "<a SETUP block and a SOURCES block - 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": "check",
  "sheet": "<a SETUP block and a SOURCES block - 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": "check",
  "sheet": "<a SETUP block and a SOURCES block - 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