km-survival-curve
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates the execution of R scripts using
Rscriptwith user-supplied parameters to perform survival analysis and generate graphical outputs. - [EXTERNAL_DOWNLOADS]: Instructions are provided to download and install standard R packages (such as
survival,survminer, andggplot2) fromhttps://cloud.r-project.org, which is the official and well-known repository for R software. - [INDIRECT_PROMPT_INJECTION]: The skill represents a potential surface for indirect prompt injection as it ingests and processes untrusted external tabular data (CSV, TXT, TSV). \n
- Ingestion points:
scripts/utils.Ruses thefreadfunction to load the file specified by the--input_fileargument. \n - Boundary markers: The script does not utilize explicit delimiters or warnings to isolate the ingested data from the agent's instructions. \n
- Capability inventory: The skill has the ability to read local files, create directories (
dir.create), write text files (writeLines), and export PDF figures. \n - Sanitization: While the script performs robust statistical validation (checking for numeric types and categorical groups), it does not sanitize the input for potential prompt injection strings that might affect subsequent agent operations.
Audit Metadata