The Agent Skills Directory

[SAFE]: The skill provides instructional guidelines for software development workflows without including executable malicious code or dangerous hidden instructions.
[EXTERNAL_DOWNLOADS]: The skill references its official repository and plugin identifiers on GitHub and platform marketplaces. These targets are well-known services and the references are used for standard installation procedures.
[PROMPT_INJECTION]: The instructions use slash commands to define agent states and tasks (e.g., /spec, /build). These are legitimate instructional patterns for structuring agent behavior and do not attempt to override the underlying model's safety guardrails or extract system prompts.
[COMMAND_EXECUTION]: While the skill lists Bash in its allowed tools, the instructions guide the agent to perform standard development tasks like git operations, testing, and file editing, which are strictly consistent with its stated purpose as a production engineering framework.

agentic-skills