The Agent Skills Directory

[SAFE]: The skill serves as a comprehensive security-focused guide for shell scripting. It explicitly instructs the agent to treat external inputs as untrusted, validate data at boundaries, and avoid practices that leak secrets (such as echoing variables or using set -x in secret-handling blocks).
[SAFE]: External tools mentioned, such as bats-core, shellcheck, and shfmt, are well-known, industry-standard utilities for testing and linting shell code. The provided CI/CD examples use standard installation methods from official registries.
[SAFE]: The skill includes advanced defensive patterns against common shell vulnerabilities, including arithmetic injection in $(( )) and delimiter-injection in /proc parsing, demonstrating a high level of security awareness and providing the agent with safe primitives.

clean-shell