The Agent Skills Directory

[SAFE]: The skill provides operational context for authoring detection content in the CrowdStrike Falcon platform. Analysis of the skill body and referenced documents revealed no malicious code, suspicious commands, or safety bypass attempts.
[PROMPT_INJECTION]: No instructions were found that attempt to manipulate the agent's behavior or bypass its core safety constraints.
[DATA_EXFILTRATION]: The skill does not access sensitive local files or perform unauthorized network communications. All query examples use benign or placeholder data.
[REMOTE_CODE_EXECUTION]: No mechanisms for downloading or running untrusted external code were identified.
[SAFE]: No hidden content, encoded payloads, or character-based evasion techniques were detected in the instructions or metadata.

crowdstrike-falcon