seo-programmatic
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and analysis of content from external data sources, which could potentially contain malicious instructions designed to manipulate the agent.\n
- Ingestion points: Assessment logic in
SKILL.mdtargets data from external CSV/JSON files, API endpoints, and database queries.\n - Boundary markers: The instructions do not specify any delimiters or warnings to disregard instructions embedded within the analyzed data records.\n
- Capability inventory: The agent is instructed to execute a local script and generate detailed SEO reports based on the ingested data.\n
- Sanitization: There are no documented procedures for sanitizing or validating the content of the data records before they are processed.\n- [COMMAND_EXECUTION]: The skill references the execution of a local utility for metadata analysis.\n
- Evidence:
SKILL.mdcontains an instruction to run\"${CLAUDE_PLUGIN_ROOT}/scripts/claude-seo\" run metadata_template.py. This utility is a vendor-provided tool associated with the author's official toolkit as indicated in the license file.
Audit Metadata