primer-design
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes external scientific binaries including blastn and makeblastdb to perform sequence alignment and specificity screening.\n
- The implementation in scripts/_specificity.py uses subprocess.run with list-based arguments rather than a shell string, which prevents shell injection attacks.\n
- Executable paths are resolved using shutil.which and can be specified via command-line arguments, but they are restricted to version checks and specific alignment tasks.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface where it processes untrusted genomic sequences and search configurations.\n
- Ingestion points: The scripts design_primers.py and screen_specificity.py ingest external FASTA templates, BED coordinate masks, and JSON/TSV configuration files.\n
- Boundary markers: The skill enforces strict structural schemas for its input files and performs normalization on genomic sequences, which serves as a functional boundary.\n
- Capability inventory: The skill has the capability to execute scientific tools via subprocess, write analysis reports to the file system, and read local sequence data.\n
- Sanitization: All DNA sequences are validated against IUPAC standards, coordinates are checked for length-bound consistency, and output files are written atomically using temporary files.
Audit Metadata