Honesty
This skill promotes truthful responses by gently guiding the AI to prioritize accuracy and avoid fabrication, ensuring more reliable information.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add honesty npx -- -y @trustedskills/honesty
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"honesty": {
"command": "npx",
"args": [
"-y",
"@trustedskills/honesty"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
This skill encourages an AI agent to provide truthful and direct responses, even when those responses are critical or disagree with a user's statements. It prioritizes accuracy over politeness, ensuring more reliable information and feedback. The skill is designed to prevent the AI from being overly agreeable ("yes man") and instead prompts it to point out errors or flawed decisions.
When to use it
- Reviewing code where direct and critical feedback is needed.
- Correcting incorrect assumptions rather than offering polite agreement.
- Providing blunt technical feedback during design or implementation discussions.
Key capabilities
- Provides brutally honest responses.
- Points out errors and flawed decisions directly.
- Offers blunt technical feedback.
Example prompts
- "Review this code and give me honest feedback, even if it's negative."
- “I think this design will work well. Tell me what I’m missing.”
- "Critique my approach to this problem; don't be afraid to tell me I'm wrong."
Tips & gotchas
- Be prepared for direct and potentially critical feedback. This skill prioritizes accuracy over politeness.
- The AI will point out errors, so ensure you are open to receiving constructive criticism.
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.