synth-manage-checks

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the gcx command-line interface via the Bash tool to perform administrative tasks on Grafana Synthetic Monitoring. This includes listing probes, creating, updating, and deleting monitoring checks. As gcx is an official tool provided by the vendor (Grafana), its use for managing vendor-specific infrastructure is consistent with the skill's stated purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection via the processing of untrusted data.
  • Ingestion points: Target URLs, hostnames, domain names, and YAML check configurations provided via user input or file reads (SKILL.md, references/check-types.md).
  • Boundary markers: The instructions rely on structured YAML templates, though specific delimiters or warnings to ignore embedded instructions within data fields are not explicitly present.
  • Capability inventory: The skill has access to Bash (for executing gcx commands), and Read, Write, and Edit tools (for managing YAML configuration files).
  • Sanitization: The gcx tool implements client-side validation to ensure that probe names, check types, and target formats conform to expected schemas before any API calls are made, significantly reducing the risk of malicious payload execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 09:04 AM
Security Audit — agent-trust-hub — synth-manage-checks