The Agent Skills Directory

[UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill installs the 'cobra' library from PyPI using pinned versioning ('0.31.1'). It also facilitates fetching metabolic models from established scientific databases (BiGG and BioModels) via the 'load_model' function. These are well-recognized and trusted academic services in the bioinformatics domain.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill requests 'Read' and 'Write' permissions to process metabolic model files and save analysis outputs. The workflows provided safely instruct the user to define a specific output directory ('OUTDIR') for all file operations, ensuring transparency.
[INDIRECT_PROMPT_INJECTION]: The skill processes structured metabolic models from SBML, JSON, and YAML files.
Ingestion points: External data is ingested through specialized IO functions ('read_sbml_model', 'load_json_model', 'load_yaml_model') in 'SKILL.md' and 'api_quick_reference.md'.
Boundary markers: No explicit delimiters are used to separate model metadata from agent instructions.
Capability inventory: The skill allows file system access ('Read', 'Write', 'Edit') and command execution ('Bash') for processing simulation data.
Sanitization: The skill relies on the standard parsing and validation logic of the 'cobra' library to handle incoming model data.

cobrapy