The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources (Adaptyv Bio API results and local CSV files) which represents a potential surface for indirect prompt injection if those sources return malicious instructions.
Ingestion points: Data is loaded from local files such as esm_candidates.csv and exp_round1_results.csv, and retrieved from the network via client.get_experiment_results() and client.list_experiments().
Boundary markers: No explicit delimiters or boundary markers are defined in the instructions to separate data from instructions during processing.
Capability inventory: The skill possesses the capability to write to the local file system (to_csv, savefig, open(output_file, 'w')) and perform network operations via the SDK to submit new experiments.
Sanitization: The skill does not demonstrate explicit validation or sanitization of fields like sequence_name or metadata before they are processed or used in terminal output.

adaptyv-bio