content-repurposer
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process long-form content from external sources, including file paths and URLs provided by the user in Step 1. This creates a surface for indirect prompt injection where malicious instructions embedded in the source content (e.g., a blog post or transcript) could potentially influence the agent's behavior.
- Ingestion points: Long-form content provided via pasted text, local file paths, or remote URLs.
- Boundary markers: The skill does not instruct the agent to use delimiters or explicitly ignore logic embedded within the source content to prevent them from being interpreted as system instructions.
- Capability inventory: The skill facilitates reading from the local file system (file paths) and the network (URLs) based on user-provided strings.
- Sanitization: No automated sanitization, validation, or escaping of the ingested content is specified, although the skill includes a mitigation where the agent presents an extraction summary to the user for verification in Step 2 before proceeding to post generation.
Audit Metadata