seo-hreflang
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its core functionality of auditing external content.
- Ingestion points: Untrusted data enters the agent context via the
/seo hreflang audit <directory-or-url>command, which fetches and processes external web pages and local files. - Boundary markers: There are no instructions or delimiters defined to isolate the audited content or to warn the agent to ignore embedded instructions within the processed data.
- Capability inventory: The skill enables the agent to fetch remote URLs, read directory contents, and perform structural analysis of HTML and metadata.
- Sanitization: The skill lacks mechanisms for sanitizing or escaping content retrieved from external sources before it is evaluated by the agent.
- [COMMAND_EXECUTION]: The skill references an external execution pattern in
references/machine-translation-qa.md(claude-seo run content_quality.py). This suggests dependency on an external script for extended functionality. While this appears to be part of the developer's broader toolset, the referenced script is not contained within the analyzed package.
Audit Metadata