accessibility-statement
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external audit reports to generate accessibility statements, which introduces an attack surface for indirect prompt injection.
- Ingestion points: In
SKILL.md, the workflow instructions state that the agent must "Check for existing audit reports" and "Map findings to known limitations." - Boundary markers: The skill lacks explicit instructions, delimiters, or warning prompts to ensure the agent ignores any instructions potentially embedded within the audit reports.
- Capability inventory: The skill includes the capability to write to the file system, specifically saving output to
ACCESSIBILITY-STATEMENT.mdas defined in the workflow steps. - Sanitization: There are no documented steps or instructions for the agent to sanitize, validate, or filter the external data gathered from audit reports before using it to generate the final document artifact.
Audit Metadata