local-content-strategy
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified. The skill is limited to providing natural language instructions and output templates for SEO analysis.
- [NO_CODE]: The skill consists entirely of markdown instructions and does not include any scripts, executables, or package dependencies.
- [INDIRECT_PROMPT_INJECTION]: The skill defines an attack surface by ingesting external data (user-provided keyword lists). The risk is negligible as the skill lacks high-risk capabilities such as network operations, file system access, or command execution.
- Ingestion points: Keyword lists provided by the user or output from preceding research skills (referenced in SKILL.md).
- Boundary markers: The skill does not explicitly define delimiters for untrusted input.
- Capability inventory: None. The skill only generates text-based strategy documents.
- Sanitization: No input sanitization or validation is implemented.
Audit Metadata