alterlab-qca
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute R scripts via Rscript for set-theoretic data analysis. This is the primary intended functionality as documented in SKILL.md and references/qca_in_r.md, and it is limited by the tool permissions defined in the skill's frontmatter.
- [INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface that reads external CSV files, which is a potential vector for indirect prompt injection.
- Ingestion points: Data is loaded in references/qca_in_r.md using the
read.csv(args[1])function. - Boundary markers: The R script template does not include specific delimiters or instructions to ignore potential commands embedded within the data.
- Capability inventory: The agent has access to shell execution via the Rscript environment as defined in the skill's configuration.
- Sanitization: No explicit validation or sanitization is performed on the input data before processing by R functions.
- [DYNAMIC_EXECUTION]: The skill instructs the agent to generate and run a local R script from a provided template.
- Evidence: Instructions in references/qca_in_r.md guide the agent to save and execute a .R script.
- Context: The script is created from a static template included in the skill's own reference files rather than from an untrusted remote source.
Audit Metadata