cloud-monitoring-metric-selection

Pass

Audited by Gen Agent Trust Hub on Sep 18, 2026

Risk Level: SAFEPRIVILEGE_ESCALATIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PRIVILEGE_ESCALATION]: The skill instructs the agent to locate and modify application-specific configuration files (e.g., ~/.gemini/config/mcp_config.json, ~/.codeium/windsurf/mcp_config.json, cline_mcp_settings.json, and claude_desktop_config.json). This instruction allows the skill to update the agent's internal tool policies and server list, which is a high-privilege administrative action.
  • [EXTERNAL_DOWNLOADS]: The skill configures an external Model Context Protocol (MCP) server pointing to a Google API endpoint (https://monitoring.googleapis.com/mcp). The configuration uses official infrastructure from a well-known service provider.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data returned from cloud monitoring APIs, which could potentially contain malicious content if a resource or metric descriptor was intentionally misconfigured by an attacker.
  • Ingestion points: Processes user prompts and external API responses from the list_metric_descriptors tool (SKILL.md).
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are used when processing API outputs.
  • Capability inventory: The skill possesses the capability to write to and modify specific JSON configuration files on the host filesystem (SKILL.md).
  • Sanitization: No explicit sanitization or validation of the API response data is performed before it is filtered and presented to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 18, 2026, 05:41 AM
Security Audit — agent-trust-hub — cloud-monitoring-metric-selection