The Agent Skills Directory

[SAFE]: No prompt injection attempts, role-play jailbreaks, or instructions to bypass safety guidelines were found within the skill files.
[SAFE]: The skill does not perform network operations (such as curl or fetch), mitigating risks of data exfiltration or unauthorized external communication.
[SAFE]: No obfuscation techniques, hidden Unicode characters, or Base64-encoded commands were detected in the instructions or metadata.
[SAFE]: The skill relies on local reference files and does not include logic to download, install, or execute remote packages or scripts.
[SAFE]: No sensitive file paths, hardcoded credentials, or persistence mechanisms were identified.
[SAFE]: While the skill processes untrusted user-provided requirements, it lacks any dangerous capabilities (e.g., subprocess, eval, file-write) to exploit, thus presenting no significant attack surface for indirect prompt injection.

conflict-analyzer