Agent skill · writing content · jeremylongshore
klingai-content-policy
Implement content policy compliance for Kling AI prompts and outputs.
Why this skill is useful
Adds a comprehensive pre-submission filtering script for Kling AI prompts that ensures compliance with content policies, preventing violations before API calls.
What it needs
Requires kling-ai account access. About 4k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.
What this skill does
Kling AI Content Policy Overview Kling AI enforces content policies server-side. Tasks with policy-violating prompts return taskstatus: "failed" with a content policy message. This skill covers pre-submission filtering to avoid wasted credits and API calls. Restricted Content Categories Kling AI prohibits prompts that generate: Category Examples ---------- --------- Violence/gore Graphic injuries, torture, weapons used violently Adult/sexual Explicit nudity, sexual acts, suggestive content Hate/discrimination Slurs, targeted harassment, supremacist imagery Illegal activity Drug manufacturing, terrorism, fraud instructions Real people Deepfakes of identifiable individuals without consent Copyrighted characters Trademarked characters (Mickey Mouse, Spider-Man) Misinformation Fake news, fabricated events presented as real Self-harm Suicide, eating disorders, self-injury instructions Pre-Submission Prompt Filter Safe Negative Prompts Always include safety-related negative prompts: Integration with Client Handling Server-Side Rejections User-Facing Guidelines When building apps with user-submitted prompts: 1. Filter before API call -- saves credits on obvious violations 2. Explain rejections clearly -- tell users what to change 3. Log violations -- track patterns for filter improvement 4. Rate limit prompt submissions -- prevent abuse 5. Review flagged content -- human review for edge cases Prerequisites A versioned policy configuration, an owner for escalation, a review queue, and a documented retention/deletion schedule. A synthetic or rights-cleared fixture set for tests. Likeness, voice, and other identifiable-person inputs require documented consent; do not rely on a prompt filter as proof of rights. A bounded credit budget and a private, watermarked draft destination. Public distribution requires a separate approval record after policy and quality checks. Instructions 1. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/klingai-content-policy