Creative Writing & Fiction Moderation

The moderation problem

Fiction legitimately contains profanity, slurs in a character's mouth, and violent or sexual scenes. The whole point of the platform is creative expression, so an over-eager filter destroys the product. But the same surface also hosts comments, DMs, and reviews where abuse is real.

Why a word list isn't enough

Word lists can't distinguish a character swearing in dialogue from a user swearing at another user. They strip the craft out of the writing and still miss targeted harassment in the surrounding social features.

How The Profanity API handles it

Use context `creative` to recognize quoting and in-character speech as legitimate, paired with `smart` mode so the skip engine spends LLM budget only on the genuinely ambiguous passages rather than every paragraph of a novel.

const res = await fetch("https://api.theprofanityapi.com/v1/check", {
  method: "POST",
  headers: {
    "Authorization": "Bearer " + process.env.PROFANITY_API_KEY,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({ text, context: "creative", mode: "smart" }),
});

const { flagged, intent, score } = await res.json();