bio-longread-alignment
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill defines command-line workflows for 'minimap2' and 'samtools'. These are standard bioinformatics utilities for processing local genomic files (FASTA, FASTQ, BAM). No arbitrary or dangerous shell command injection surfaces were identified.
- [DATA_EXPOSURE] (SAFE): Analysis confirms that the skill only interacts with scientific data files provided by the user. No sensitive paths (e.g., ~/.ssh, ~/.aws) or environmental variables are accessed.
- [REMOTE_CODE_EXECUTION] (SAFE): There are no patterns involving the download or execution of remote scripts (no curl/wget piped to bash). All operations rely on local binaries.
- [PROMPT_INJECTION] (SAFE): The skill body consists of technical instructions and examples. There are no attempts to override system prompts, bypass safety filters, or use 'DAN'-style jailbreak patterns.
- [INDIRECT_PROMPT_INJECTION] (SAFE): 1. Ingestion points: Reads sequence data from FASTQ/FASTA files and file lists via 'cat'. 2. Boundary markers: Absent, as these are binary file formats. 3. Capability inventory: Execution of minimap2, samtools, and awk for data processing. 4. Sanitization: Standard tool-level parsing of structured biological data. Evaluation: The risk is negligible as the ingested data (genomic sequences) is processed by specialized binary parsers that do not interpret natural language instructions.
Audit Metadata