The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill implements machine learning model persistence using the joblib library. \n
Evidence: SKILL.md contains the load_model function which uses model = joblib.load(filepath). \n
Risk: joblib utilizes pickle serialization under the hood. Deserializing untrusted files can lead to arbitrary code execution if a user is induced to load a maliciously crafted model file. \n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external construction project data. \n
Ingestion points: instructions.md defines the ingestion of historical datasets in CSV or Excel format containing project features and costs. \n
Boundary markers: The instructions lack explicit guidance for the agent to use delimiters or to ignore potential instructions embedded within data fields. \n
Capability inventory: The skill utilizes filesystem permissions to read datasets and write model files. \n
Sanitization: There is no evidence of input validation or sanitization for the data values processed by the ML scripts.

cost-prediction