Skills
skills/pjt222/agent-almanac/add-rcpp-integration/Gen Agent Trust Hub

add-rcpp-integration

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates standard R package development workflows and does not contain malicious patterns.\n- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute chmod +x cleanup. This is a standard and required step for CRAN-compliant packages to ensure the cleanup script can run during the build process. The script content is limited to deleting temporary compilation artifacts (.o, .so, .dll).\n- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes existing package code.\n
  • Ingestion points: The skill reads the existing package's DESCRIPTION file and src/ directory content during setup and compilation (usethis::use_rcpp, Rcpp::compileAttributes).\n
  • Boundary markers: Absent; standard for this type of development-focused tool.\n
  • Capability inventory: The agent uses Write, Edit, and Bash to modify files and run build tools like devtools::load_all().\n
  • Sanitization: Not applicable as the skill is intended for use on developer-controlled source code.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 11:57 PM