check-approve
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to prompt injection via user-supplied parameters that are interpolated into shell commands.
- Ingestion points: Parameters such as
<REQ-CODE>,--be-container, and--userdefined inSKILL.mdare directly accepted from the user. - Boundary markers: Absent. There are no delimiters or instructions to treat these inputs as literal data or to ignore embedded control characters.
- Capability inventory: The skill has the capability to generate shell commands and write them to the local file system (e.g.,
scratchpad/check-approve-<REQ-CODE>.sh). - Sanitization: Absent. The skill does not perform any escaping or validation on user inputs, allowing potential command injection sequences (e.g.,
;,&&,|) to be passed into the final command template. - [DYNAMIC_EXECUTION]: The skill generates shell script files (
.sh) at runtime based on user input. While categorized as simple script generation from templates, the lack of input validation means the resulting executable content is not guaranteed to be safe for the user to execute.
Audit Metadata