analyzing-tiktok-hashtag-performance
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from TikTok, including video descriptions, captions, and creator metadata. Findings: 1. Ingestion points: External data enters through the Apify actor's dataset output as described in SKILL.md. 2. Boundary markers: Absent. The skill does not define specific delimiters or instructions to ignore embedded commands in the scraped content. 3. Capability inventory: The skill can write output to files (CSV/JSON) and execute shell commands via node and curl. 4. Sanitization: Absent. There is no explicit sanitization step described for the external content before it is processed or displayed.
- [DYNAMIC_EXECUTION]: The skill exposes a customMapFunction parameter in SKILL.md which accepts a raw JavaScript function string to transform data on the Apify platform. Providing executable code as an input string is a security-sensitive pattern that could be abused if an attacker influences the input.
- [COMMAND_EXECUTION]: The skill utilizes shell commands (node scripts/run_actor.js, curl) to interact with the Apify service. While it correctly suggests using environment variables ($APIFY_TOKEN) instead of hardcoding secrets, command-line interpolation of user-provided parameters like hashtags or keywords into shell commands is a potential injection risk if parameters are not properly sanitized or escaped.
Audit Metadata