The Agent Skills Directory

[SAFE]: The skill's primary focus is on reproducible data visualization and quality assurance. No malicious patterns, such as credential exfiltration, obfuscation, or unauthorized network access, were detected.
[COMMAND_EXECUTION]: The script scripts/test_helper.py uses subprocess.run to invoke the skill's own helper.py script. This is used strictly for unit testing the validation logic and does not allow for arbitrary command injection or execution of untrusted external code.
[INDIRECT_PROMPT_INJECTION]: The skill defines a workflow that ingests external data (JSON, CSV, or text) to generate visual reports. While this creates a theoretical attack surface, it is a standard part of the skill's functionality.
Ingestion points: Measurement results and structured data artifacts ingested in SKILL.md.
Boundary markers: None explicitly defined in the instructions for input data.
Capability inventory: The skill possesses file-write capabilities (saving GIFs, contact sheets, and JSON reports) and local command execution for tests.
Sanitization: Input data is processed through the Pillow image library, which acts as a buffer against direct instruction execution by the model.

animated-infographic