skills/brycewang-stanford/awesome-agent-skills-for-empirical-research/stata-data-cleaning/Gen Agent Trust Hub
stata-data-cleaning
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill suggests installing community-developed Stata packages using the standard
ssc installcommand. - Evidence: Recommends installing
unique,mdesc, andlabutilfrom the Statistical Software Components (SSC) archive, which is the official community repository for Stata. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process data from external, untrusted sources such as APIs and survey datasets.
- Ingestion points: External data enters the agent's context through the Stata
usecommand and variable inspection steps outlined in the cleaning pipeline instructions inSKILL.md. - Boundary markers: The instructions do not define clear boundary markers or safety guidelines to ignore potential instructions embedded within the data values or metadata of the source files.
- Capability inventory: The skill generates code for data transformation, variable labeling, and logic validation using
assertstatements based on the processed data. - Sanitization: The instructions lack explicit sanitization steps to filter out or escape potentially malicious strings found within the raw data files.
Audit Metadata