paper-spine-humanize
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external academic manuscripts, creating an attack surface where hidden instructions in the source text could influence the agent's behavior. \n
- Ingestion points: The skill reads academic text provided as input, and the validation script
scripts/humanize_check.pyscans all.texfiles within thefinal_paperdirectory. \n - Boundary markers: Absent. The instructions do not specify delimiters or provide warnings for the agent to ignore instructions that might be embedded within the input manuscripts. \n
- Capability inventory: The associated script
scripts/humanize_check.pyhas the ability to read files and write reports to the project's output directory, which could be exploited if an injection causes the agent to mismanage output paths. \n - Sanitization: Absent. The manuscript content is analyzed and processed as raw text without filtering for potential prompt injection triggers or suspicious patterns.
Audit Metadata