pcp-engine
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes an 'Audit' mode which processes user-provided copy or URLs to suggest improvements. This creates a surface for indirect prompt injection if the analyzed content contains malicious instructions.
- Ingestion points: User-supplied URLs or copy ingested during the 'Step 0' phase in SKILL.md and used in 'Audit Mode' in refinement-and-audit.md.
- Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the ingested copy.
- Capability inventory: The skill is limited to text analysis and generation; it does not define or request tools for file system writing, network exfiltration, or shell command execution.
- Sanitization: No sanitization or filtering logic is specified for the input data.
- [SAFE]: The skill instructions and reference materials focus entirely on marketing and behavioral psychology. There are no patterns suggesting data exfiltration, persistence, or remote code execution. No obfuscated strings or URLs were found across the skill files.
Audit Metadata