sgrna-design-guide

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions and a comprehensive list of URLs to download CRISPR design datasets from the Broad Institute's CRISPick service (portals.broadinstitute.org). These downloads originate from a well-known research institution and are necessary for genome-scale CRISPR experiment design.
  • [COMMAND_EXECUTION]: Documentation includes shell command examples for downloading (wget) and decompressing (gunzip) biological datasets, as well as searching within files (grep). These are standard utilities for managing scientific data.
  • [PROMPT_INJECTION]: The skill ingests data from a bundled CSV file (resource/addgene_grna_sequences.csv) containing biological metadata and validated sequences. This ingestion of structured data represents a potential surface for indirect prompt injection, although the provided content is benign.
  • Ingestion points: The Option 1.1 search method uses pandas.read_csv to load the curated database of sgRNA sequences.
  • Boundary markers: The skill does not implement specific boundary markers or 'ignore' instructions when processing the CSV data.
  • Capability inventory: The skill has the capability to execute shell commands (wget, gunzip) and perform file system operations.
  • Sanitization: The skill does not perform sanitization or schema validation on the CSV content before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 03:35 PM
Security Audit — agent-trust-hub — sgrna-design-guide