bio-sashimi-plots
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of local bioinformatics command-line tools such as ggsashimi.py, rmats2sashimiplot, and voila. These operations are implemented in examples/plot_sashimi.py using Python's subprocess.run with command arguments passed as a list, which is a standard security practice that avoids shell injection vulnerabilities.
- [INDIRECT_PROMPT_INJECTION]: The skill includes logic to process external scientific data files, such as rMATS output files (SE.MATS.JC.txt), to parameterize the visualization commands. -- Ingestion points: The batch_plot_rmats_events function in examples/plot_sashimi.py reads data from user-supplied file paths. -- Boundary markers: The skill does not use specific delimiters or instructions to ignore instructions embedded within the genomic data files. -- Capability inventory: The skill uses subprocess.run to invoke external CLI tools for data plotting. -- Sanitization: The skill utilizes list-based command construction for all subprocess calls, which inherently prevents command injection from potentially malicious data in the input files.
Audit Metadata