Skills
Navigate
skills/k-dense-ai/scientific-agent-skills/waypoint-bio/Gen Agent Trust Hub

waypoint-bio

Fail

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: CRITICAL
Full Analysis
  • [SAFE]: The skill provides integration for the waypoint-bio Python package and Hugging Face repositories.
  • [EXTERNAL_DOWNLOADS]: Downloads data from Hugging Face Hub (e.g., outpost-bio/Waypoint-6m, outpost-bio/Atlas). These are trusted resources belonging to a well-known service and the model's publisher.
  • [REMOTE_CODE_EXECUTION]: The skill uses trust_remote_code=True when loading the Hugging Face tokenizer. This is standard and required behavior for this specific architecture's custom tokenizer. Users are advised in references/python-api.md to pin a revision hash to ensure the integrity of the remote code.
  • [COMMAND_EXECUTION]: The skill provides several Python scripts (scripts/profiler_to_waypoint.py, scripts/vocab_coverage.py) for local data processing. Analysis of these scripts shows standard data manipulation using pandas and argparse with no evidence of malicious command injection or shell escapes.
  • [CREDENTIALS_SAFE]: The skill requires a Hugging Face token (HF_TOKEN) for access to gated repositories. Instructions correctly guide users to use standard authentication methods (hf auth login) or environment variables, which is a safe practice for credential management.
Recommendations
  • CRITICAL: 1 file(s) identified as malware by FileRep - DO NOT USE
  • Contains 1 malicious URL(s) - DO NOT USE
Audit Metadata
Risk Level
CRITICAL
Analyzed
Sep 3, 2026, 03:30 AM
Security Audit — agent-trust-hub — waypoint-bio