The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external Benchling tenants which could contain malicious instructions designed to influence the agent.
Ingestion points: Data is retrieved from Benchling registry entities, notebook entries, and event payloads via the Python SDK and REST API as described in references/core_capabilities.md and references/api_endpoints.md.
Boundary markers: The implementation snippets do not demonstrate the use of boundary markers or instructions to ignore embedded commands in the retrieved data.
Capability inventory: The skill is granted allowed-tools: Read Write Edit Bash, and Python examples demonstrate file writing and network operations as shown in SKILL.md and references/core_capabilities.md.
Sanitization: No explicit sanitization or content validation of Benchling API responses is evidenced in the skill's reference code.

benchling-integration