skills.rest
Omar-Obando
Omar-ObandoCOMMUNITY
Explore all skills by this author →
GitHub

anti-hallucination-techniques

Verify AI-generated claims by evaluating source credibility and evidence quality.

44|9|Updated May 7, 2026
Search Tags:#fact-checking#technical documentation#source verification#hallucination detection#citation audit#evidence-based reasoning
One-click install
npx skills add https://github.com/Omar-Obando/qwen-orchestrator --skill anti-hallucination-techniques
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: anti-hallucination-techniques
Source: https://github.com/Omar-Obando/qwen-orchestrator/tree/main/skills/anti-hallucination-techniques
Command: npx skills add https://github.com/Omar-Obando/qwen-orchestrator --skill anti-hallucination-techniques

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps stop AI from presenting incorrect or fabricated information by forcing claims to be verified against credible sources and evidence.

Core Features & Use Cases

  • Source credibility verification: Evaluate the currency, relevance, expertise, diversity, independence, and evidence quality of sources (via CREDIBLE-style checks).
  • Fact-checking and validation frameworks: Apply claim validation steps (find evidence, evaluate strength, check counter-evidence, and state uncertainty).
  • Hallucination pattern detection: Identify common failure modes like fake citations, incorrect technical details, and unsupported statistics.

Example use: When an AI generates an answer about an API endpoint, version, or parameter list, this Skill guides you to verify the information with authoritative documentation and to flag unsupported claims or fake references.

Quick Start

Use the anti-hallucination-techniques skill to verify whether the following AI claim is supported by credible evidence: “The Qwen Code CLI supports feature X on version Y with parameter Z…”

Frequently Asked Questions about anti-hallucination-techniques

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

FAQPage Schema
How do I detect AI hallucinations in technical documentation?▼

Detect AI hallucinations by identifying unsupported statistics, fake citations, and incorrect technical details, then validating them against credible sources and evaluating evidence strength.

What is the best way to fact-check AI generated content for fake citations?▼

Fact-checking AI generated content requires applying validation frameworks that find evidence, evaluate source credibility, check for counter-evidence, and explicitly state any uncertainty regarding factual assertions.

How does source verification work for evidence-based reasoning?▼

Source verification evaluates the currency, relevance, expertise, diversity, and independence of sources to ensure evidence-based reasoning is grounded in high-quality, credible documentation.

Can I use this to audit code examples and API parameters?▼

Yes, you can audit code examples and API parameters by verifying AI-generated claims about specific endpoints or versions against authoritative technical documentation to identify unsupported assertions.

What are common AI hallucination patterns in generated statistics?▼

Common AI hallucination patterns in generated statistics include fabricated data points, fake references, and unsupported factual assertions that lack credible evidence or proper citation audit trails.

Related Skills

Explore More Education & Research Skills→
bacchus-labsbacchus-labs
4

researching-web-sources

Research web topics via strategic searches and source verification.

Official
Intermediate
q734738781q734738781
13

citation-grounding-and-claim-discipline

Review manuscript text for unsupported claims and missing citations.

Community
Intermediate
LAI-YEN-CHUNLAI-YEN-CHUN
1

fact-checker

Evaluate claims by sourcing evidence and assigning credibility verdicts.

Community
Advanced
ShubhamsabooShubhamsaboo
131.7k

fact-checker

Guide structured fact-checking through claim identification, evidence gathering, and credibility assessment.

Community
Basic