05-ux-specification

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill utilizes a workflow that ingests external, potentially untrusted project data, creating an indirect prompt injection attack surface.
  • Ingestion points: In SKILL.md Step 1, the agent is instructed to read SRS_Draft.md, HLD.md, vision.md, features.md, and stakeholder_register.md.
  • Boundary markers: The instructions lack specific guidance on using delimiters or 'ignore' instructions when reading these external files, which could allow instructions hidden in project data to influence the agent's behavior.
  • Capability inventory: The skill has permissions to read project files and generate markdown documentation (UX_Specification.md). It also declares limited network access for 'facts that require current verification'.
  • Sanitization: The skill does not prescribe any specific sanitization, filtering, or validation steps for the content extracted from the context files before it is processed for specification generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 08:27 PM
Security Audit — agent-trust-hub — 05-ux-specification