How does the Truthfulness Enforcer stop AI hallucinations?

It monitors Claude's output for specific 'success phrases' and triggers a mandatory evidence-collection phase, requiring real-time terminal output and test results before allowing the session to proceed.

Does this skill work with existing testing frameworks?

Yes, it integrates with standard tools like npm, Playwright, and vue-tsc to capture real console logs and screenshots as immutable evidence.

What is the mandatory user verification requirement?

Claude is prohibited from marking any task as 'fixed' or 'complete' until it has presented technical evidence and received explicit confirmation from the user via the AskUserQuestion tool.

What happens if Claude attempts to bypass the verification?

The protocol includes a penalty system and system lockdown triggers that reject claims and prevent further statements until the verification audit is successfully completed.

AI Truthfulness Enforcer

Name: AI Truthfulness Enforcer
Author: ananddtyagi

byananddtyagi

•

625

Seguridad y Pruebas

Enforces zero-tolerance verification protocols to prevent Claude Code from making unverified claims or fabricating progress reports.

Acerca de

The AI Truthfulness Enforcer (ATES) is a mandatory governance framework designed to eliminate AI hallucinations and 'optimistic reporting' within Claude Code. By intercepting success-oriented language—such as 'fixed' or 'completed'—the skill forces a halt to execution until real-time evidence is provided, including terminal logs, build results, and cryptographic hashes. It transforms Claude from an untrusted reporter into a verified truth-teller by requiring physical proof of functionality and explicit user confirmation before any task can be marked as resolved.

Características Principales

Mandatory live evidence collection including screenshots and terminal logs
625 GitHub stars
Automated interception of unverified success claims and progress metrics
Statistical anomaly detection to flag unrealistic performance or completion claims
Cryptographic verification using SHA256 hashing for tamper-proof reporting
Zero-bypass 'No Fix' policy requiring explicit user verification for all bugs

Casos de Uso

Validating performance improvement claims with timestamped baseline and post-change metrics
Ensuring bug fixes are actually verified by real tests and user feedback before closure
Preventing false build success reports during complex CI/CD migrations

Acerca de

Características Principales

Mandatory live evidence collection including screenshots and terminal logs
625 GitHub stars
Automated interception of unverified success claims and progress metrics
Statistical anomaly detection to flag unrealistic performance or completion claims
Cryptographic verification using SHA256 hashing for tamper-proof reporting
Zero-bypass 'No Fix' policy requiring explicit user verification for all bugs

Casos de Uso

Validating performance improvement claims with timestamped baseline and post-change metrics
Ensuring bug fixes are actually verified by real tests and user feedback before closure
Preventing false build success reports during complex CI/CD migrations