The Agent Skills Directory

[NO_CODE]: The skill consists entirely of markdown documentation (SKILL.md) and a static YAML configuration (agents/openai.yaml). There are no scripts, binaries, or automated tasks to analyze for execution-based threats.
[SAFE]: The content defines a logical framework for process analysis based on industry standards like Lean and Six Sigma. The instructions are focused on improving semantic fidelity and structural integrity of user-provided workflows.
[INDIRECT_PROMPT_INJECTION]: While the skill is designed to process untrusted user input (workflows), its primary output is a structured audit report or a refined process. It lacks capabilities like file writing, network access, or command execution that would allow an indirect injection to escalate into a technical compromise.

improve-processes