opentrons-integration
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides templates and instructions for generating Python code that controls physical laboratory hardware (Opentrons Flex and OT-2 robots). This creates an attack surface where malicious instructions embedded in external experimental data (e.g., a CSV of pipetting volumes or a project description) could influence the agent to generate protocols that damage hardware or misuse hazardous reagents.
- Ingestion points: The skill is designed to be used by an agent to interpolate experimental parameters into the
run()function structure defined inSKILL.mdand the templates in thescripts/directory. - Boundary markers: The provided templates do not include delimiters or system-level instructions to ignore potential commands embedded within processed experimental data.
- Capability inventory: The skill grants the agent the capability to control liquid handling (pipettes), temperature modules, heater-shakers, and thermocyclers as documented in
references/api_reference.md. - Sanitization: There is no evidence of input validation or sanitization logic to ensure that parameters sourced from external data do not contain logic that overrides the intended protocol flow.
Audit Metadata