yt-content-planner
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the user and passes it to other automated skills, creating a vulnerability where malicious instructions in the input could influence the agent's behavior or output.
- Ingestion points: The skill collects a channel niche, audience descriptions, content pillars, and recent video titles/view counts from the user (SKILL.md, Step 1). It also processes content from user-provided channel URLs.
- Boundary markers: The instructions lack explicit delimiters or "ignore instructions" wrappers when passing these user-provided inputs to sub-skills like
yt-title-optimizeroryt-hook-scripter. - Capability inventory: The skill has the ability to parse external URLs using
lib/url_parser.pyand publish content vialib.publish(...)(SKILL.md, Steps 1 & 8). - Sanitization: There is no mention of sanitization, filtering, or validation for the user-provided titles or the content parsed from external URLs before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill invokes local library scripts and functions to handle data processing and automation tasks.
- Evidence: The implementation references
lib/url_parser.pyfor URL analysis in Step 1 andlib.publish(...)for executing publication tasks in Step 8.
Audit Metadata