The Agent Skills Directory

[SAFE]: The skill is a wrapper for the flowio library, a standard tool for handling flow cytometry data. Analysis of the instructions and provided documentation confirms that all functionality is focused on scientific data processing. No malicious patterns such as prompt injection, data exfiltration, or obfuscation were detected.
[EXTERNAL_DOWNLOADS]: The skill requires the installation of the flowio package via uv pip install. This is a legitimate package available on official registries and is consistent with the skill's stated purpose.
[COMMAND_EXECUTION]: The skill includes instructions for package installation. These are standard administrative tasks for setting up a Python environment and do not involve suspicious or privileged command execution.
[INDIRECT_PROMPT_INJECTION]: The skill processes external FCS files, which represent an ingestion point for untrusted data. While this creates a theoretical attack surface, the functionality is restricted to data extraction and transformation into structured numeric formats (NumPy arrays) or CSV files for scientific analysis. No evidence of unsafe interpolation of this data into agent prompts was found.
Ingestion points: FlowData('sample.fcs') in SKILL.md and references/api_reference.md.
Boundary markers: None explicitly defined in the prompt instructions.
Capability inventory: File reading (FlowData), file writing (create_fcs, write_fcs), and data transformation (NumPy/Pandas).
Sanitization: None described; however, the data is primarily processed as binary scientific data rather than natural language.

flowio