Format numbers
Fail
Audited by Gen Agent Trust Hub on Feb 15, 2026
Risk Level: HIGHPROMPT_INJECTIONNO_CODE
Full Analysis
- [Indirect Prompt Injection] (HIGH): The skill operates on untrusted external data (TSX files) and has the capability to modify that data. This creates a surface for indirect prompt injection where malicious instructions could be embedded in code comments or strings to manipulate the agent's behavior during an automated edit session. * Ingestion points: File content matching the glob pattern **/*.tsx. * Capability inventory: Modification of source code through the agent's fileEdit capability. * Boundary markers: Absent. * Sanitization: Absent.
- [No Code] (SAFE): This skill contains markdown-based instructions only and does not include executable shell scripts or Python binaries, limiting the risk of direct malicious code execution from the skill itself.
Recommendations
- AI detected serious security threats
Audit Metadata