bio-atac-seq-atac-qc
Pass
Audited by Gen Agent Trust Hub on Apr 2, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: Python snippets in
SKILL.md(such as theatac_qc_reportfunction) usesubprocess.check_output(..., shell=True)with direct string interpolation of variables like{bam_file}and{peaks_file}. This pattern is susceptible to command injection if filenames contain shell metacharacters.\n- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection through malicious file naming, as untrusted input paths are used in security-sensitive shell operations.\n - Ingestion points: File paths (
bam_file,peaks_file,tss_bed,genes.gtf) are used as inputs for functions inSKILL.mdandexamples/atac_qc_metrics.R.\n - Boundary markers: None; input paths are interpolated directly into shell strings without delimiters.\n
- Capability inventory: Use of
subprocess.check_output(shell access),pysam(file analysis), andopen()(file access) throughout the skill code.\n - Sanitization: No validation or sanitization of input filenames is performed.
Audit Metadata