Skills
skills/williamzujkowski/standards/coding-standards/Gen Agent Trust Hub

coding-standards

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (HIGH): The file shell/scripts/install-shell-tools.sh downloads binary executables and source code from non-whitelisted GitHub repositories.
  • Evidence: Functions install_shellcheck_binary, install_shfmt_binary, and install_bats_from_source download content from github.com/koalaman, github.com/mvdan, and github.com/bats-core respectively.
  • [REMOTE_CODE_EXECUTION] (HIGH): The script shell/scripts/install-shell-tools.sh executes setup scripts downloaded from the internet with elevated privileges.
  • Evidence: install_bats_from_source clones the bats-core repository and executes sudo ./install.sh directly.
  • [COMMAND_EXECUTION] (HIGH): Several scripts utilize sudo to perform system-level modifications based on external data.
  • Evidence: shell/scripts/install-shell-tools.sh uses sudo mv, sudo chmod, and sudo ./install.sh to install tools into /usr/local/bin.
  • [EXTERNAL_DOWNLOADS] (LOW): The python/resources/configs/.pre-commit-config.yaml file depends on multiple external GitHub repositories to fetch and run pre-commit hooks at runtime.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill setup scripts ingest user-provided project names and package names without sanitization, potentially allowing local file system manipulation if an attacker provides a path-traversal string as a project name.
  • Ingestion points: PROJECT_NAME argument in swift/scripts/setup-swift-project.sh, python/scripts/setup-project.sh, etc.
  • Boundary markers: Absent.
  • Capability inventory: Scripts can create directories, write files, and initialize git repositories.
  • Sanitization: None detected.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:19 PM