The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the user to install the benchling-sdk package via standard package managers like pip (uv) or poetry. These are references to official, well-known libraries for interacting with the Benchling platform.
[COMMAND_EXECUTION]: The documentation provides example shell commands for installing dependencies (uv pip install benchling-sdk, poetry add benchling-sdk). These are benign and standard for setting up a development environment.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from Benchling's registry and notebook entries, which constitutes an external attack surface where untrusted data could theoretically influence agent behavior.
Ingestion points: Data is ingested through SDK methods such as benchling.dna_sequences.list(), benchling.containers.list(), and benchling.entries.list() in the SKILL.md file.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the processing logic.
Capability inventory: The skill possesses network communication capabilities (via the Benchling SDK) and file-writing capabilities (e.g., creating sequences.csv in the Data Export example).
Sanitization: No explicit sanitization or validation of data retrieved from the Benchling platform is documented before processing or exporting.

benchling-integration