content-research-writer
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for basic local environment setup using standard shell commands (
mkdir,cd,touch) within the user's home directory. These are used for routine file organization and do not involve sensitive paths or high-privilege operations. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted user-provided writing drafts and research materials, which constitutes a vulnerability surface.
- Ingestion points: User-provided drafts (e.g.,
article-draft.md) and research materials read into the agent's context. - Boundary markers: Absent; the instructions do not specify the use of delimiters or 'ignore embedded instructions' markers for user-provided text.
- Capability inventory: File system access (reading drafts, writing outlines) and web-based research capabilities.
- Sanitization: Absent; there are no instructions to sanitize or validate the content of ingested files before processing.
Audit Metadata