check-prd
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data including issue bodies, comments, and repository content to validate execution readiness.\n
- Ingestion points: Accesses live issue data and repository revisions as defined in the 'Inputs' section of
SKILL.mdand the 'Canonical Storage and Identity' section ofreferences/execution-readiness.md.\n - Boundary markers: Utilizes structured templates, specific headings (e.g.,
## Implementation Plan), and machine-readable headers to delimit untrusted content from control logic.\n - Capability inventory: Operations are restricted to read-only inspection of repositories and issue trackers; no capabilities for file modification, network transmission, or arbitrary command execution are present in this skill.\n
- Sanitization: Implements a strict validation process including EARS grammar checks for acceptance criteria and a SHA256 digest requirement to ensure the integrity of the requirements body.
Audit Metadata