agm-in-a-box
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or dangerous commands were detected. The skill consists entirely of instructional Markdown and text templates.
- [INDIRECT_PROMPT_INJECTION]: The skill defines an ingestion surface for untrusted data by asking users to provide their organization's constitution or rules.
- Ingestion points: The 'Required Inputs' section in SKILL.md asks for user-pasted rules and documents.
- Boundary markers: Absent; the instructions do not specify delimiters for the pasted content.
- Capability inventory: None; the skill's functionality is limited to producing Markdown text. It has no access to shell commands, network tools, or file system modifications.
- Sanitization: Absent; the skill relies on the AI agent's internal safety filters for processing user input.
Audit Metadata