diagnose
Pass
Audited by Gen Agent Trust Hub on May 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were identified. The included shell script is a benign template for human-in-the-loop interaction.
- [COMMAND_EXECUTION]: The skill is designed to execute a variety of diagnostic commands, including tests, CLI tools, and browser automation scripts (Playwright/Puppeteer), which are essential for its bug-reproduction purpose.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes external data such as network traces, event logs, and bug reports. 1. Ingestion points: Phase 1 (network requests, payloads, event logs) and Phase 2 (user failure descriptions). 2. Boundary markers: Absent; there are no specific instructions to use delimiters or ignore instructions within external data. 3. Capability inventory: Significant capabilities including CLI command execution, test running, and browser automation. 4. Sanitization: Absent; the skill does not specify validation or sanitization of ingested artifacts.
Audit Metadata