alterlab-proteinmpnn
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes PDB structure files, representing an external data ingestion surface. Any skill processing external files may be susceptible to indirect instructions embedded in that data.\n
- Ingestion points: The skill parses backbone PDB files into JSONL format as described in
SKILL.md.\n - Boundary markers: No explicit instructions or delimiters to ignore embedded content were found.\n
- Capability inventory: The skill uses
Bash(python:*)andBash(uv:*)to run scientific design scripts.\n - Sanitization: The skill relies on standard scientific parsing scripts without explicit security validation of the PDB content.\n- [EXTERNAL_DOWNLOADS]: The skill references and utilizes code from the
dauparas/ProteinMPNNrepository on GitHub for its sequence design tasks.\n- [COMMAND_EXECUTION]: The skill uses Python and theuvtool to execute modeling and parsing scripts on the local filesystem as part of the protein design workflow.
Audit Metadata