tt-repurposer
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from various external and untrusted sources (LinkedIn posts, blogs, newsletters, X threads) which could contain hidden instructions intended to override agent behavior.
- Ingestion points: External content provided via prompts or links as described in Step 1 of the processing flow in SKILL.md.
- Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" markers when handling source text.
- Capability inventory: The skill has the ability to publish content to TikTok via the lib.publish tool as defined in Step 12.
- Sanitization: While the skill uses a "Humanizer pass" (Step 10) to scrub AI-like vocabulary, it lacks technical sanitization to filter or escape potential injection patterns in the input text.
- Mitigation: The workflow includes an "Approval card" (Step 11) that requires user confirmation before the "On approval" publishing step (Step 12), providing a human-in-the-loop safeguard.
Audit Metadata