landing-page-copy
Warn
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: MEDIUMMETADATA_POISONINGINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The 'Skill Boundaries' section contains descriptions that are entirely inconsistent with the skill's stated purpose, specifically referencing 'audio production workflows' and 'audio engineering expertise' instead of landing page copy. Additionally, there is a discrepancy between the author listed in the frontmatter ('ClawFu') and the internal metadata block ('MKTG Skills').
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to audit external landing pages via URLs or pasted text, creating a vulnerability surface where malicious instructions embedded in a target page could influence the agent.
- Ingestion points: 'Audit an existing page' section in SKILL.md.
- Boundary markers: None; the instructions do not provide delimiters to separate external data from system instructions.
- Capability inventory: The skill contains markdown instructions only and does not include scripts with command execution, network, or file system access capabilities.
- Sanitization: No input validation or content filtering mechanisms are defined for the audited content.
Audit Metadata