bids

Pass

Audited by Gen Agent Trust Hub on May 18, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The script scripts/update_schema.py fetches the latest BIDS schema and Extension Proposals (BEPs) from official sources (bids-specification.readthedocs.io and the bids-standard GitHub organization). These are trusted community resources for the neuroscience domain.
  • [COMMAND_EXECUTION]: The skill provides standard instructions for using neuroimaging CLI tools like heudiconv, dcm2bids, and bids-validator. These are used for data transformation and validation according to scientific standards.
  • [INDIRECT_PROMPT_INJECTION]: As the skill processes externally provided datasets (JSON sidecars, TSV phenotypic data, and directory structures), there is a theoretical surface for indirect prompt injection. However, the risk is negligible as the skill uses standard scientific libraries for parsing and the data typically originates from controlled research environments.
  • Ingestion points: Reads BIDS directory structures and metadata files (dataset_description.json, *.json, *.tsv) using pybids and bids-validator.
  • Boundary markers: Not explicitly defined in logic, but standard BIDS parsing libraries enforce strict schema validation.
  • Capability inventory: File system reads/writes for dataset organization; execution of standard neuroimaging CLI tools.
  • Sanitization: Relies on pybids and json/csv standard library parsers which include built-in sanitization for their respective formats.
Audit Metadata
Risk Level
SAFE
Analyzed
May 18, 2026, 03:45 AM
Security Audit — agent-trust-hub — bids