The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes external libraries to interact with the Benchling platform and process sequence data.
Evidence: The skill instructs the user to install the benchling-sdk package using uv add benchling-sdk or uv pip install benchling-sdk. It also references the biopython library through from Bio import SeqIO in its usage examples.
[INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes diverse data types from the Benchling environment, such as DNA sequences, registry entities, and electronic lab notebook (ELN) entries. This creates a surface for indirect prompt injection if the content within Benchling is malicious.
Ingestion points: The skill ingests data via benchling.dna_sequences.list(), benchling.entries.get_entry_by_id(), and benchling.workflow_tasks.list() across SKILL.md and references/sdk_reference.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided templates for handling retrieved data.
Capability inventory: The skill has the ability to write to the file system (Write, Edit) and execute shell commands (Bash(python:*), Bash(curl:*)).
Sanitization: There is no evidence of content sanitization or validation performed on data retrieved from Benchling before it is used in subsequent operations.

alterlab-benchling