The Agent Skills Directory

[SAFE]: The skill's scripts and documentation were thoroughly reviewed, and no malicious patterns, obfuscation, or security vulnerabilities were detected.
[DATA_EXFILTRATION]: The Python scripts (scripts/agent.py, scripts/process.py) perform local statistical analysis on network logs. They do not initiate unauthorized network connections or attempt to exfiltrate data.
[REMOTE_CODE_EXECUTION]: The skill uses standard Python libraries for processing and does not involve remote code downloads, dynamic execution, or unauthorized package installations.
[PROMPT_INJECTION]: The provided markdown instructions and metadata do not contain any patterns attempting to override agent behavior or bypass safety constraints.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted network log data, representing a Category 8 surface. However, it lack functions for executing shell commands or performing network operations based on that data, mitigating injection risks.
Ingestion points: Network log parsing in scripts/agent.py and scripts/process.py.
Boundary markers: None.
Capability inventory: Limited to local file read/write and mathematical operations; no subprocess or network capabilities.
Sanitization: Standard type conversion and field mapping during log ingestion.

hunting-for-beaconing-with-frequency-analysis