infrahub-reporting-issues
Pass
Audited by Gen Agent Trust Hub on May 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
Bashtool to run several discovery commands such aspip show,ansible-galaxy collection list,helm list, andinfrahubctl --version. These are used to collect version information for the bug report. - [EXTERNAL_DOWNLOADS]: The skill fetches issue templates directly from official GitHub repositories within the
opsmillorganization using thegh api. The content is decoded and presented to the user as a template. - [PROMPT_INJECTION]: There is a surface for indirect prompt injection because the skill incorporates user-provided descriptions and external tool outputs (like GitHub search results) into the issue body. This is mitigated by a mandatory human review step before any submission occurs.
- [DATA_EXFILTRATION]: While the skill's function is to send data externally to GitHub, it includes a robust sanitization framework in
rules/environment-info-sanitization.md. This rule explicitly instructs the agent to redact internal IP addresses, hostnames, filesystem paths, and secrets, ensuring that only non-sensitive environment data is shared.
Audit Metadata