The Agent Skills Directory

EXTERNAL_DOWNLOADS (MEDIUM): The skill instructions and technique guides (e.g., techniques/02_survey_resampling.md, techniques/03_synthetic_control.md) suggest using the ssc install command to download and install third-party Stata packages like reghdfe, boottest, ritest, oaxaca, and synth. These packages originate from the Statistical Software Components (SSC) archive, which is a community repository and constitutes an unverifiable external dependency.
COMMAND_EXECUTION (MEDIUM): The skill's operational core involves generating Stata .do scripts and executing them via system shell commands (e.g., stata -e do filename.do). Although this is the primary intended use case for the skill, the ability to execute arbitrary scripts in the host environment is a high-risk capability.
PROMPT_INJECTION (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) due to its data ingestion capabilities. 1. Ingestion points: The skill reads external data files using import delimited and import excel commands in techniques/00_data_prep.md. 2. Boundary markers: There are no specific boundary markers or safety instructions in the code templates to prevent the agent from interpreting instructions embedded within the user's datasets. 3. Capability inventory: The skill has the ability to execute shell commands, modify the file system, and access the network via package managers. 4. Sanitization: No data validation or sanitization logic is provided to inspect data content before it is processed by the analysis pipeline.

stata-analyst