Install
openclaw skills install tiktok-content-creation-complianceReview TikTok Shop seller and creator promotional content compliance using official TikTok Shop Seller University guidance. Use when an AI agent needs to check or revise a TikTok Shop video, LIVE, script, caption, hook, storyboard, exported video file, listing image, product claim, giveaway language, or AI-generated ad; decide whether content is permitted, risky, or prohibited; and catch misleading claims, product-listing mismatch, originality problems, AI disclosure issues, prohibited editing tactics, health or wellness claims, giveaway or gambling issues, IP risks, platform-safety issues, or affiliate creator eligibility limits.
openclaw skills install tiktok-content-creation-complianceClassify TikTok Shop promotional content as Permitted, Risky - revise, Prohibited, or Unclear - verify live policy. This skill applies to both sellers promoting their own products and creators promoting products for others, including affiliate creator workflows.
Use the official policy summary in references/policy-summary.md as the starting point, and stay conservative when a claim, visual, product anchor, promotion mechanic, or edit could mislead users.
If the user asks for the latest or current rule, reopen the official URLs in the reference file before giving a final answer. TikTok Shop policy guidance changes over time.
Use references/policy-summary.md as the canonical rule map, not as a page-by-page dump.
When multiple TikTok Shop pages repeat the same idea:
Unclear - verify live policy if the conflict is still unresolvedPractical order of authority:
Determine what you are reviewing:
Also determine the promotion context:
If the user only provides a rough idea, infer the missing structure and review the idea anyway.
If the user provides a video file, inspect the actual media instead of relying only on the script or filename. Prefer:
Review in this order:
Load references/policy-summary.md before deciding.
If the user is working from a creator education page, treat that page as a shortcut into the relevant canonical rule. Do not overcount duplicate warnings just because the same issue appears in several training pages.
When reviewing a finished video, explicitly check:
Use exactly one primary verdict:
PermittedRisky - reviseProhibitedUnclear - verify live policyFor every flagged issue, name:
When revising content:
Do not claim content is safe if the policy signal is mixed. Mark it Risky - revise or Unclear - verify live policy.
Return reviews in this structure:
Verdict: <Permitted | Risky - revise | Prohibited | Unclear - verify live policy>
Why:
- <plain-English summary>
Flags:
- <issue> -> <policy reason + source>
Safer rewrite:
- <replacement wording or scene direction>
Open questions:
- <only if something cannot be resolved from the supplied content>