content-calendar
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a template-driven planning tool. It does not perform network requests, access sensitive system credentials, or execute arbitrary code. The stated purpose aligns with its functional behavior.
- [INDIRECT_PROMPT_INJECTION]: The skill reads from a local file
.agents/youtube-context.mdto inform its planning logic. - Ingestion points:
.agents/youtube-context.md(accessed in the 'Before Starting' section). - Boundary markers: Absent. The skill treats the content of the file as trusted context.
- Capability inventory: No dangerous tools are requested or used; the skill lacks subprocess execution, network exfiltration, or file system writing capabilities.
- Sanitization: None. Because the skill's only output is planning text and tables, the risk of a successful injection leading to system compromise is negligible.
Audit Metadata