x-writer-codex
Pass
Audited by Gen Agent Trust Hub on Sep 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from the
context/directory to generate social media content, creating a vulnerability surface where malicious instructions in those files could influence agent behavior. - Ingestion points: Reads brand, calendar, best performers, and compliance files from the
context/directory. - Boundary markers: No delimiters or "ignore instructions" wrappers are used when referencing context data.
- Capability inventory: File system read access for context files and write access for outputting drafts to
outputs/x/. - Sanitization: No validation or filtering logic is specified for the content retrieved from the input files.
Audit Metadata