skill-comply
Tests whether an AI agent actually follows its rules by running it through tricky scenarios and reporting what it does.
Installation
Paste this into Claude Code, Cursor, or any agent that can run commands.
What this skill does
What it does:
- Checks if an AI agent actually follows its written rules and skills
- Creates test scenarios with different levels of strictness
- Runs the agent and records every tool it uses
- Compares what the agent actually did to what it should have done
- Creates a report showing how well the agent followed the rules
When to use it:
- After writing new rules or skills to see if the agent really follows them
- When you want to check if an agent works correctly even when instructions are unclear or conflicting
- During regular maintenance to make sure nothing broke
- When you add new rules and want to verify the agent actually uses them
The full skill text could not be loaded right now. It installs fine with the command above, or read it at the source.
Mirrored from the author's public source. Install counts from the open skills registry.