landing-page-editor
Pass
Audited by Gen Agent Trust Hub on May 18, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
astriaCLI to fetch and update workspace data. Access is restricted to theastrianamespace within the Bash tool, which is a standard security practice to limit the scope of execution to the vendor's own tools. - [PROMPT_INJECTION]: The skill processes existing HTML content fetched from workspaces. This provides a surface for indirect prompt injection, as malicious instructions could be embedded in the HTML to influence the agent's behavior during the editing process.
- Ingestion points: Landing page HTML is fetched from the server and stored in
/tmp/landing_page.htmlfor processing (SKILL.md). - Boundary markers: None present; the agent is instructed to edit the file directly without specific delimiters or warnings to ignore content.
- Capability inventory: The skill has access to Bash (scoped to
astria:*), Read, Write, and Edit tools across the filesystem (SKILL.md). - Sanitization: No explicit sanitization or validation of the fetched HTML is performed before the agent interacts with it.
Audit Metadata