seo-fix-apply
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from plan and manifest files to execute changes. This vulnerability is mitigated by a multi-step confirmation protocol.
- Ingestion points: Data is ingested from
plan.jsonandmanifest.jsonlocated within the run directory (SKILL.md). - Boundary markers: The protocol explicitly restricts operations to a specific list of
confirmed_change_idsprovided by the user. - Capability inventory: The agent can invoke
Edit,Write, and shell commands via external scripts to modify project files and remote resources. - Sanitization: The skill mandates the redaction of credential-shaped data from all logs.
- [DYNAMIC_EXECUTION]: The skill executes external Node.js adapter scripts located in the plugin's internal directory. The specific script to be executed is determined at runtime based on the target platform (e.g., Shopify, HubSpot).
- Evidence: Logic uses
<SCRIPTS_DIR>/adapters/<adapter>.mjsto dynamically load and run platform-specific logic.
Audit Metadata