Guidance
Provides helpful suggestions and context-aware prompts to steer conversations toward desired outcomes, boosting user engagement and clarity.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add guidance npx -- -y @trustedskills/guidance
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"guidance": {
"command": "npx",
"args": [
"-y",
"@trustedskills/guidance"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
Guidance Skill Guide
What it does
The guidance skill provides structured, step-by-step instructions to AI agents, helping them navigate complex tasks with clarity and precision. It ensures that agents follow specific workflows or adhere to defined constraints during execution.
When to use it
- Breaking down multi-step coding projects into manageable phases for an agent.
- Enforcing strict safety protocols or compliance rules during automated operations.
- Guiding agents through unfamiliar APIs or proprietary system architectures.
- Providing context-aware direction when an agent encounters ambiguous user requests.
Key capabilities
- Delivers clear, sequential instructions tailored to task complexity.
- Supports constraint enforcement for consistent output quality.
- Adapts guidance dynamically based on real-time agent progress.
- Integrates seamlessly with Claude Code templates for enhanced workflow control.
Example prompts
- "Use the guidance skill to outline a three-step process for debugging a Python script."
- "Provide step-by-step instructions for deploying a Docker container while adhering to security best practices."
- "Guide me through setting up a local development environment with specific version requirements."
Tips & gotchas
Ensure you provide sufficient context upfront so the guidance remains relevant throughout the task. Avoid overly vague instructions, as they may lead to misaligned agent behavior or incomplete workflows.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.