specialist-testing-testing-reality-checker

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external sources such as 'QA findings' and 'system claims', which constitutes a vulnerability surface for indirect prompt injection.
  • Ingestion points: Data enters the agent context through 'QA findings', 'claimed features', and 'previous agent' reports mentioned in SKILL.md.
  • Boundary markers: Absent; the skill does not specify delimiters or instructions to ignore embedded commands within the ingested content.
  • Capability inventory: The subagent prompt in SKILL.md implies access to 'automated tools' for screenshots and 'reality check commands'.
  • Sanitization: Absent; there are no instructions to sanitize, escape, or validate the findings from other agents or tools.
  • [PROMPT_INJECTION]: The subagent instructions in SKILL.md use strong directives such as 'Stop Fantasy Approvals' and 'Default to FAILED'. These are evaluated as domain-specific persona constraints for a testing role rather than malicious overrides of the system's core safety filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 08:50 PM
Security Audit — agent-trust-hub — specialist-testing-testing-reality-checker