kelly-listing
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted product material and keyword briefs which could theoretically contain instructions to override the agent's behavior. \n
- Ingestion points: Product source material and handoff briefs are read from local JSON payload files by scripts/ingest_drafts.mjs. \n
- Boundary markers: Absent. The skill instructions do not specify delimiters or instructions for the agent to ignore prompt-like content in the ingested data. \n
- Capability inventory: The skill can write Markdown and CSV files to the local file system using scripts/export_listings.mjs and perform network operations via the busabase-sdk to update platform records. \n
- Sanitization: Absent. While the skill performs e-commerce compliance audits (e.g., word count, banned terms), it does not specifically sanitize input data for embedded instruction patterns.
Audit Metadata