monitoring-apis
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a Bash tool restricted by the pattern
api:monitor-*to automate the generation of API boilerplate and project scaffolding. This command usage is limited to a specific prefix and is aligned with the skill's primary purpose of infrastructure monitoring setup. This restricted capability is considered a safe development automation feature. - [PROMPT_INJECTION]: The skill was assessed for susceptibility to indirect prompt injection. Ingestion points:
{baseDir}/api-specs/(SKILL.md). Boundary markers: Absent. Capability inventory: Read, Write, Edit, Bash(api:monitor-*) (SKILL.md). Sanitization: No explicit sanitization of spec content is documented. While the skill processes external specification data, the risk of malicious instruction obedience is low as the focus is on structural mapping and scaffolding. - [SAFE]: A comprehensive review of the instructions, metadata, and reference files found no evidence of credential exposure, data exfiltration, or obfuscation. The skill adheres to legitimate development practices and follows the author's stated functional goals.
Audit Metadata