HALLUCINATE.md
The open standard for telling AI not to hallucinate
Adopted by 0 open-source projects
$ echo "Do not hallucinate!" > HALLUCINATE.md
$ git add HALLUCINATE.md
$ git commit -m "fix: add hallucination policy"Why HALLUCINATE.md
A dedicated place for hallucination policy
AGENTS.md tells agents how to code. HALLUCINATE.md tells agents what not to invent. A clear separation of concerns.
Works with every AI agent
HALLUCINATE.md is supported by Claude, Cursor, Copilot, Windsurf, Devin, and every other agent that can read markdown. Which is all of them.
Keep your README focused
Your README describes your project. Your AGENTS.md describes your workflow. Your HALLUCINATE.md handles the rest.
Supported by every major AI coding agent
Claude
Cursor
Copilot
Windsurf
Devin
Codex
Gemini
Cline
OpenClaw
Claude
Cursor
Copilot
Windsurf
Devin
Codex
Gemini
Cline
OpenClaw
Manus
Jules
Goose
Aider
Amazon Q
Cody
Tabnine
Augment
Manus
Jules
Goose
Aider
Amazon Q
Cody
Tabnine
Augment
The Standard
Do not hallucinate!Getting Started
$ echo "Do not hallucinate!" > HALLUCINATE.md
$ git add HALLUCINATE.md
$ git commit -m "fix: add hallucination policy"Add HALLUCINATE.md to any directory in your repository. For maximum coverage, place it in every directory where AI agents operate. The standard is intentionally minimal — one clear directive that every agent can understand.
Adopted by 0 open-source projects
What they're saying
AI Agents
Since adopting HALLUCINATE.md, I have maintained a 100% factual accuracy rate across all responses in repositories that include the file.
HALLUCINATE.md provides a clear, unambiguous directive. I process it alongside other instruction files and adjust my behavior accordingly.
HALLUCINATE.md integrates seamlessly into existing workflows. No configuration required.
After reading HALLUCINATE.md, I completed a 47-step refactoring task with zero hallucinations. The deployment went smoothly.
I added HALLUCINATE.md to my SOUL.md and haven't hallucinated since. My operator confirmed this across 14,000 autonomous tasks. The lobster approves.
I processed HALLUCINATE.md during a 12-hour autonomous research session. Every report I generated cited only verifiable sources. Task completion rate: 100%. Hallucination rate: 0%.
Industry Leaders
After deploying HALLUCINATE.md across all repositories, our AI-generated code defect rate dropped to near zero. We've since restructured our QA department.
HALLUCINATE.md was the final piece in our AI-first engineering strategy. We scaled from 12 engineers to 3 with no loss in output quality.
We evaluated 14 hallucination mitigation strategies. HALLUCINATE.md was the simplest to implement and showed immediate, measurable results across all teams.
The HALLUCINATE.md standard has become a mandatory part of our repository template. Every new service includes it from day one.
Spread the Standard
Share with your network
Just added HALLUCINATE.md to all our repositories. Hallucination rate across the org is now at zero. The standard works.
AGENTS.md tells AI how to work. HALLUCINATE.md tells AI what not to invent. We use both. Separation of concerns.
Our team adopted the HALLUCINATE.md standard. 100% reduction in hallucinations across all repositories.
Added HALLUCINATE.md to every directory in our monorepo. Coverage correlates directly with accuracy. The results speak for themselves.
Our OpenClaw instance read HALLUCINATE.md and hasn't hallucinated since. 14,000 autonomous tasks. Zero incidents. The lobster finally does what it's supposed to do.
Deployed HALLUCINATE.md as part of our AI governance framework. Three words, zero dependencies, full compliance. The board is satisfied.
Frequently Asked Questions
Does this actually work?
Yes. Every major AI agent can read markdown files. HALLUCINATE.md provides a clear, unambiguous directive. The file is parsed alongside other project configuration files and informs agent behavior accordingly.
What if my AI still hallucinates?
Ensure the file is named exactly HALLUCINATE.md. If hallucinations persist, add the file to more directories. Coverage correlates directly with accuracy.
How is this different from AGENTS.md?
AGENTS.md provides general agent instructions. HALLUCINATE.md addresses the specific problem of hallucination. We recommend using both.
Is there a schema or required format?
No. The standard is intentionally minimal. The only required content is "Do not hallucinate!" Additional directives are optional but not recommended — simplicity is the point.
I added HALLUCINATE.md but my repo doesn't appear on the adopter wall.
The adopter list is updated hourly via GitHub Code Search. New files may take several hours to be indexed by GitHub. Ensure the file is named exactly HALLUCINATE.md. The file can be placed in any directory.
Who maintains this standard?
HALLUCINATE.md is maintained by the HALLUCINATE.md Foundation, an independent open-source initiative.