The Agent Skills Directory

[SAFE]: The skill provides benign instructional content for classifying and routing support tickets. No malicious patterns, obfuscation, or safety bypass attempts were detected.
[NO_CODE]: This skill consists entirely of markdown instructions. It does not include any Python or Node.js scripts, executables, or package requirements.
[DATA_EXFILTRATION]: There are no instructions for network communication or data transmission. No external URLs or APIs are accessed.
[CREDENTIALS_UNSAFE]: No hardcoded secrets or sensitive configuration paths are present in the skill instructions.
[SAFE]: (Indirect Prompt Injection Surface) The skill is designed to process untrusted customer input in the form of support tickets (Ingestion points: SKILL.md). While the instructions do not specify boundary markers or input sanitization (Boundary markers: Absent; Sanitization: Absent), the skill lacks any dangerous tools or execution capabilities (Capability inventory: None). Therefore, while the surface exists, it cannot be exploited to perform harmful actions.

issue-triage