hallucination-detector

Validate technology choices against the project tech stack and codebase.

Updated Jan 5, 2026
One-click install
npx skills add https://github.com/Sjdjdiejdrirhdkjej/Claude-code-conductor --skill hallucination-detector-sjdjdiejdrirhdkjej
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: hallucination-detector
Source: https://github.com/Sjdjdiejdrirhdkjej/Claude-code-conductor/tree/main/.claude/skills/hallucination-detector
Command: npx skills add https://github.com/Sjdjdiejdrirhdkjej/Claude-code-conductor --skill hallucination-detector-sjdjdiejdrirhdkjej

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The hallucination-detector prevents hallucinated technical decisions during feature work by validating all technology choices, API suggestions, and database/schema changes against the project's documented tech stack and existing codebase.

Core Features & Use Cases

  • Validates technology suggestions against docs/project/tech-stack.md
  • Verifies entities, APIs, and packages exist in codebase or documentation
  • Blocks non-conforming suggestions and provides corrected alternatives with evidence

Quick Start

  1. Load the project's tech stack from docs/project/tech-stack.md and initialize the detector
  2. Detect and classify a suggestion (e.g., "Use Redux for state management"), compare against documented choices, and block if mismatched
  3. Provide a corrected recommendation with citations to evidence sources

Frequently Asked Questions about hallucination-detector

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I validate technology choices against my project's tech stack during development?

You validate technology choices by loading docs/project/tech-stack.md as the single source of truth and cross-checking suggestions against the existing codebase and documentation to block non-conforming decisions.

What is a tech hallucination in software engineering and how do I prevent it?

A tech hallucination is a non-conforming technical decision suggesting APIs, packages, or schemas that do not exist in your codebase. Prevent it by requiring evidence and validating all technology suggestions against your documented tech stack.

How do I stop AI from suggesting APIs and database schemas that don't exist in my codebase?

Stop invalid AI suggestions by using a validation mechanism that verifies entities, APIs, and packages against your existing codebase and documentation, blocking suggestions that lack evidence from these sources.

Can I use automated validation to correct bad technical suggestions during the Spec to Plan to Implement cycle?

Yes, automated validation operates during the Spec, Plan, and Implement cycles to detect mismatched suggestions, block them, and provide corrected recommendations with citations to evidence sources.

Does this tech stack validation require any external dependencies to run?

No, this tech stack validation requires no external dependencies to operate. It relies entirely on loading your project's documented tech stack and analyzing the existing codebase to enforce evidence-based decisions.

What is the best way to ensure code review suggestions match my project's documented architecture?

The best way is to enforce automated evidence-based validation that cross-checks code review suggestions against your documented tech stack and codebase existence, blocking mismatches and providing corrected alternatives.