The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and review external Bash scripts, shebangs, and CI workflow configurations, which creates a surface for indirect prompt injection where malicious instructions could be embedded in the code under review.
Ingestion points: Processes .sh files, shebang headers, and CI workflow files (shell: bash).
Boundary markers: The instructions do not specify delimiters or safety prompts to isolate the reviewed content from the agent's internal logic.
Capability inventory: The skill utilizes command execution for development tools and local orchestration scripts.
Sanitization: There is no evidence of sanitization or filtering applied to the external content before analysis.
[COMMAND_EXECUTION]: The skill directs the agent to execute specific shell commands for task orchestration and quality assurance.
Evidence: Mentions execution of python3 scripts/resolve_style_guides.py and python3 scripts/validate_trigger_matrix_sync.py for skill management.
Evidence: References several standard CLI tools such as shellcheck, shfmt, bats, ruff, and cargo for performing automated quality gates.

bash-style-guide