The Agent Skills Directory

[COMMAND_EXECUTION]: Th e skill directs the agent to execute internal Python scripts (scripts/resolve_style_guides.py and scripts/validate_trigger_matrix_sync.py) to manage trigger logic and ensure consistency within the skill set.
[COMMAND_EXECUTION]: Th e skill includes a Quality Gate Command Matrix which specifies numerous commands for the agent to execute, including package managers and test runners like npm, cargo, dotnet, ruff, and pytest.
[PROMPT_INJECTION]: Th e skill is vulnerable to indirect prompt injection as it processes and refactors untrusted code and workflow files, which could contain malicious instructions designed to influence the agent's behavior.
Inge stion points: Th e skill ingests files matching various extensions and shebangs defined in the SKILL.md description and the trigger-matrix.md (e.g., .zsh, .py, .js, .go, .rs, .tf).
Boun dary markers: Th e skill does not implement delimiters or specific instructions to the agent to treat the code content as non-instructional data.
Cap ab i li ty inventory: Th e agent has the capability to run a wide range of system commands and language toolchains as defined in the quality gate matrix.
Sa n i tization: Th ere are no explicit requirements or steps for the agent to sanitize or validate the content of the scripts before processing them for review or execution.

zsh-style-guide