SPACE-prd-writer

Pass

Audited by Gen Agent Trust Hub on Mar 31, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [NO_CODE]: The skill consists exclusively of markdown instructions and a PRD template. It does not contain executable code, scripts, or automation logic that interacts with the host system.
  • [DATA_EXFILTRATION]: No network-enabled tools or external URLs are referenced for data transmission. All processing occurs within the agent's context using the provided templates.
  • [COMMAND_EXECUTION]: There are no shell commands, system calls, or privilege escalation patterns (like sudo or chmod) within the instruction set.
  • [PROMPT_INJECTION]: The instructions are strictly focused on the persona of a Product Manager and the logical steps of document creation. No safety bypasses, role-play jailbreaks, or instruction-override patterns were detected.
  • [REMOTE_CODE_EXECUTION]: The skill does not perform any external downloads or package installations. It relies solely on the internal model logic and the provided references/prd-template.md file.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 31, 2026, 04:59 PM