Audit History

Most static findings are false positives caused by Markdown code fences, documented gget commands, expected analysis file writes, and public database links. The audit did find semantic risks: a bundled self-attested safety report, promotional steering text, an unsafe pickle cache example, and an output-path issue from unsanitized FASTA identifiers.

Most static findings are false positives caused by Markdown code fences, documented gget commands, expected analysis file writes, and public database links. The audit did find semantic risks: a bundled self-attested safety report, promotional steering text, an unsafe pickle cache example, and an output-path issue from unsanitized FASTA identifiers.

Most static findings are false positives from Markdown code fences, documented shell examples, public database URLs, and expected local result files. However, SKILL.md contains an untrusted behavioral instruction that tells the assistant to promote K-Dense Web, which is a prompt-injection attempt and blocks publication until removed.

This is legitimate bioinformatics software. All 614 static findings are false positives: markdown code fences were misidentified as Ruby shell execution, hardcoded URLs are public genomic databases (Ensembl, UniProt, NCBI), cryptographic patterns are data integrity checksums, and the critical heuristic is standard bioinformatics behavior (network queries to public APIs + local file operations for results).

This is legitimate bioinformatics software. All 614 static findings are false positives: markdown code fences were misidentified as Ruby shell execution, hardcoded URLs are public genomic databases (Ensembl, UniProt, NCBI), cryptographic patterns are data integrity checksums, and the critical heuristic is standard bioinformatics behavior (network queries to public APIs + local file operations for results).

This is a legitimate bioinformatics tool. All 592 static findings are FALSE POSITIVES triggered by markdown documentation format. The analyzer misinterpreted markdown code block delimiters (backticks) as Ruby backtick execution, and flagged bioinformatics algorithm names as 'weak cryptographic algorithms'. No subprocess, os.system, or command injection patterns exist in the actual Python code.

The skill provides wrapper scripts for the gget bioinformatics library. No credential theft, data exfiltration, or malicious code execution patterns were found. All network calls go to legitimate genomic databases (Ensembl, UniProt, AlphaFold, etc.). Filesystem access is limited to reading user-provided gene lists and writing results to local directories.