astrofy-internal-links
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes HTML build output and MDX source files to identify link destinations and suggest improvements. These files are untrusted data sources that could contain malicious instructions designed to override agent behavior during the analysis or editing phase.
- Ingestion points: The skill reads HTML files, MDX source files, and the
.astrofy/config/paths.jsonconfiguration file. - Boundary markers: Instructions mention using Markdown/MDX parsers for delimited editing and preserving structured elements (frontmatter, imports, code), which provides structural boundaries but does not prevent instruction injection within the text content.
- Capability inventory: The skill uses
astrofy links scanfor data ingestion and has the capability to write/edit MDX files. - Sanitization: The instructions do not specify sanitization or escaping of the content read from files before it is processed by the agent.
- [COMMAND_EXECUTION]: The skill executes the
astrofyCLI tool (astrofy links scan --json) to generate diagnostics and link indices. This command is executed locally against the project's build artifacts to facilitate the link management workflow.
Audit Metadata