social-media-manager
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill instructions and associated code are consistent with the stated purpose of social media management. No evidence of data exfiltration, credential theft, or unauthorized command execution was found.- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing external data by reading files such as
marketing-context.mdfor brand voice and audience personas. However, the skill lacks exploitable capabilities, rendering this surface benign. - Ingestion points: Reads
marketing-context.md(SKILL.md) to gather brand and audience context. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands in the context file.
- Capability inventory: The included script
scripts/social_calendar_generator.pyis limited to standard Python library functions (datetime, json, argparse) and does not possess network, file-writing, or system execution capabilities. - Sanitization: There is no explicit sanitization of the input context data before it is used by the agent.
Audit Metadata