multi-service-debug
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides numerous bash workflows utilizing
gcloud,vercel, andcurlto inspect service configurations, check IAM policies, and fetch logs. These tools represent a powerful interface for interacting with production environments during troubleshooting sessions. - [EXTERNAL_DOWNLOADS]: The documentation recommends the installation of third-party debugging utilities via
brewandnpm. Additionally, it includes a command to executewscatusingnpx, which involves fetching and running code from the npm registry at runtime. - [PROMPT_INJECTION]: The skill facilitates the ingestion of external log data which creates a surface for indirect prompt injection. An attacker could craft specific log messages designed to influence the agent's reasoning when it processes those logs.
- Ingestion points: Service logs fetched via
gcloud logging readandvercel logsinSKILL.md. - Boundary markers: Absent. The instructions do not specify delimiters to isolate untrusted log content from the agent's core instructions.
- Capability inventory: Access to cloud management CLIs (
gcloud,vercel), network request utilities (curl), and WebSocket debuggers (websocat,wscat). - Sanitization: Absent. There are no steps provided to sanitize or validate the content of the logs before they are analyzed by the agent.
Audit Metadata