The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves ingesting and processing external datasets, which introduces a vulnerability surface where data could contain hidden instructions intended to influence the agent's behavior.
Ingestion points: The skill utilizes pd.read_csv, pd.read_json, and pd.read_sql in SKILL.md to load data from external files and databases.
Boundary markers: No specific delimiters or instructions are provided to the agent to treat the ingested data as untrusted or to ignore embedded natural language instructions.
Capability inventory: The skill possesses the capability to read local files and execute SQL queries via the Pandas library as shown in SKILL.md.
Sanitization: The provided examples do not include sanitization, filtering, or validation steps to ensure that the content of the data files does not interfere with the agent's logic.

data-wrangling