grid-creator
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the execution of the 'dxs' command-line interface and 'jq' to retrieve, validate, and update grid configurations. Evidence includes commands like 'dxs configuration validate', 'dxs configuration upsert', and 'jq .json envelope.json'. These operations are consistent with the skill's stated purpose for developers.
- [INDIRECT_PROMPT_INJECTION]: The skill processes grid configuration data fetched from external sources using 'dxs configuration get'. This creates an attack surface where a maliciously crafted configuration could attempt to influence the agent's behavior during the editing process.
- Ingestion point: The skill reads JSON configuration bodies retrieved via the 'dxs' tool.
- Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the fetched JSON data as untrusted or to ignore embedded commands.
- Capability inventory: The agent possesses the capability to execute shell commands ('dxs') and perform file system operations.
- Sanitization: There is no evidence of sanitization or validation of the input data specifically designed to prevent prompt injection before processing.
Audit Metadata