The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands for mutation testing depending on the project type, such as cargo mutants, npx stryker run, mutmut run, and mix muzak. These are legitimate commands for its stated purpose of software testing.
[EXTERNAL_DOWNLOADS]: The skill provides instructions for installing mutation testing dependencies from official registries via cargo install, npm install, and pip install. The tools mentioned—including Stryker, mutmut, and cargo-mutants—are widely recognized industry-standard utilities.
[PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection as it processes untrusted data from the local project environment and tool outputs to determine testing outcomes.
Ingestion points: Project configuration files (e.g., Cargo.toml, package.json, pyproject.toml), source code files, and stdout/stderr from mutation testing tools.
Boundary markers: Not present; the skill lacks explicit instructions to treat tool output as untrusted or to ignore embedded instructions within processed files.
Capability inventory: The skill allows the agent to execute shell commands and modify code based on the interpreted results of the mutation run.
Sanitization: No validation or sanitization of tool output or file content is specified before the data is interpreted by the agent.
[NO_CODE]: The skill consists solely of a markdown-based instruction file and metadata; it does not include any accompanying scripts, binaries, or executable code assets.

mutation-testing