The Agent Skills Directory

[SAFE]: The skill is a non-executable informational document providing a change management framework. It contains no shell commands, scripts, or automated network operations.
[DATA_EXPOSURE]: No hardcoded credentials, tokens, or sensitive file paths (such as SSH keys or environment files) were found in the skill content.
[REMOTE_CODE_EXECUTION]: The skill does not perform any external downloads, package installations (pip/npm), or dynamic code execution.
[PROMPT_INJECTION]: No instructions were found that attempt to bypass AI safety filters or override the underlying system prompt.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The sub-commands /think:kotter, /kotter:assess, and /kotter:diagnose ingest user-supplied organization context and change initiative descriptions (SKILL.md).
Boundary markers: The skill does not explicitly define delimiters for untrusted user input.
Capability inventory: The agent utilizes standard text-processing tools (Read, Write, Edit, Agent) to analyze the input and provide structured feedback.
Sanitization: No specific sanitization or filtering logic is present, but the scope is limited to business consulting advice, posing minimal risk.

kotter