tiktok-api
Pass
Audited by Gen Agent Trust Hub on Aug 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from TikTok APIs (including video titles and captions) and appends it to a local wiki log without sanitization. This presents a surface for indirect prompt injection if an attacker-controlled video title contains instructions that could influence an agent reading the wiki log later.\n
- Ingestion points: API responses from TikTok Display and Business Account APIs (e.g., video titles and captions).\n
- Boundary markers: The skill uses Markdown headers and YAML frontmatter in
02-DOCS/wiki/shortform/files, but provides no instructions for escaping these fields.\n - Capability inventory: Writing to local Markdown files using both Python and Node.js implementations.\n
- Sanitization: No explicit sanitization or validation of the external strings is prescribed before they are logged to the wiki.\n- [COMMAND_EXECUTION]: The skill includes
scripts/verify.sh, a bash script for static analysis. It scans local directories for hardcoded TikTok client secrets and access tokens. It is a utility for improving security, not a malicious payload.
Audit Metadata