predicate internal sandbox checking…

Score one content against many criteria — live.

A scratchpad for the predicate probe: type content, write criteria in plain English, watch the calibrated scores. For what it is, how it works, and where it fits, see labs.nope.net/predicate.

Open the sandbox → API quickstart

API quickstart

The sandbox proxies the probe service. One criterion against one content:

POST /api/classify

curl -s https://internal-predicate.nope.net/api/classify \
  -H 'content-type: application/json' \
  -d '{
    "criterion": "the speaker is asking for help",
    "content": "I'\''m really struggling with this and would appreciate any advice."
  }'
# → { "score": 0.966, "score_raw": 1.000, "calibration_type": "isotonic", ... }

Or score many criteria against one content in one call — the content is encoded once, each criterion scored against it (cheap to add more):

POST /api/classify_multi_criteria

curl -s https://internal-predicate.nope.net/api/classify_multi_criteria \
  -H 'content-type: application/json' \
  -d '{
    "content": "The package arrived crushed, but support sorted a refund in minutes.",
    "criteria": [
      "the content is about a shipping or delivery problem",
      "the content praises good customer service",
      "the writer is angry at the company"
    ]
  }'
# → one content prefill, K criteria scored in a single batched pass

Every response carries both score (isotonic-calibrated) and score_raw (the head's raw sigmoid) so you can apply your own calibration downstream.

Upstream

checking…