datadog-review-dashboard
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Shelltool to run thepupCLI with user-provided variables like<service-name>and<dashboard-id>. This pattern is susceptible to command injection if the agent fails to sanitize or escape input before passing it to the shell environment. - [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it ingests and processes external dashboard definitions which are not fully under the agent's control.
- Ingestion points: Dashboard definitions fetched via
pup dashboards getin theSKILL.mdworkflow. - Boundary markers: Absent; the skill does not specify the use of delimiters or 'ignore' instructions when handling the dashboard data.
- Capability inventory:
Shell(command execution),Read(file system access), andAskUserQuestion(interaction). - Sanitization: Absent; there is no mention of validating or filtering the content of the dashboard JSON/output before the agent audits it.
Audit Metadata