Superpowers
An agentic skills framework & software development methodology that works.
Anti-hallucination guidelines for Claude Code — teach AI to say "I don't know" instead of guessing.
> "Hallucinations are hard to anticipate, hard to catch, and the wrong answer often looks exactly like it could be the right one." — Anthropic A set of behavioral guidelines to reduce AI hallucinations in Claude Code, derived from Anthropic's official explanation of why AI models hallucinate and practical tactics to catch and prevent it. AI models work by predicting the next word based on patterns learned from
git clone --depth 1 https://github.com/instantX-research/anthropic-anti-hallucinate-skills.git ~/.claude/skills/anthropic-anti-hallucinate-skills
Restart Claude Code afterwards. See the README for skill-specific setup.
Using Codex CLI? Copy the same folder to ~/.codex/skills/anthropic-anti-hallucinate-skills instead.
No specific triggers listed.
> "Hallucinations are hard to anticipate, hard to catch, and the wrong answer often looks exactly like it could be the right one." — Anthropic A set of behavioral guidelines to reduce AI hallucinations in Claude Code, derived from Anthropic's official explanation of why AI models hallucinate and practical tactics to catch and prevent it. AI models work by predicting the next word based on patterns learned from
Clone it into your personal skills folder: git clone --depth 1 https://github.com/instantX-research/anthropic-anti-hallucinate-skills.git ~/.claude/skills/anthropic-anti-hallucinate-skills — then restart Claude Code.
Anthropic Anti Hallucinate Skills is open source on GitHub (16 stars at last check) and published by instantX-research.
Browse other development skills in the Development category of this directory.