skills.rest
chrono-meta
chrono-metaCOMMUNITY
Explore all skills by this author →
GitHub

phantom-quench

Back-trace verifiable claims in AI-generated artifacts to declared sources.

7|Updated May 26, 2026
Search Tags:#citation checking#artifact audit#source grounding#phantom claims#test case validation#ai content verification
One-click install
npx skills add https://github.com/chrono-meta/forge-harness --skill phantom-quench
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: phantom-quench
Source: https://github.com/chrono-meta/forge-harness/tree/main/plugins/fh-meta/skills/phantom-quench
Command: npx skills add https://github.com/chrono-meta/forge-harness --skill phantom-quench

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

AI-generated artifacts like test cases, analysis reports, and design documents often contain claims that are not actually present in the project's source materials — these unfounded "phantom claims" come from the model's training data rather than the actual project sources, leading to incorrect tests, invalid documentation, and faulty decision-making.

Core Features & Use Cases

  • Automated Claim Extraction: Pulls all verifiable claims (proper nouns, numerical values, branching conditions, state transitions) from artifacts for back-tracing.
  • Source Back-Tracing: Greps declared local source files to confirm claims exist in the correct context, and fetches external citations to verify they actually support the claimed fact.
  • Severity Classification & Prescriptions: Grades ungrounded claims by risk (S/A/B) and provides specific fix steps, plus detects structural generation issues like source files not being read during artifact creation.
  • Use Case: A QA team using AI to generate test cases from a requirements document can run this audit to catch test steps that reference non-existent API endpoints, incorrect threshold values, or unsupported external citations before the test cases are added to the suite.

Quick Start

Use the phantom-quench skill to audit the test case file 'user_flow_tc.md' against its declared source files 'product_requirements.md' and 'api_spec_v2.yaml' to identify any unfounded claims.

Frequently Asked Questions about phantom-quench

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

FAQPage Schema
How do I check AI-generated test cases for phantom claims not present in source documents?▼

To check AI-generated test cases for phantom claims, you can audit the artifact by back-tracing every verifiable claim to its declared local source files using grep and verifying external citations via web fetch. This process detects unfounded statements originating from model training data rather than actual project sources.

What are phantom claims in AI-generated technical documentation and analysis reports?▼

Phantom claims in AI-generated technical documentation are unfounded assertions originating from model training data rather than declared project sources. They lead to incorrect tests and invalid documentation by stating facts, numerical values, or branching conditions not actually present in source materials.

How do I validate that AI-generated design documents are grounded in local source files?▼

You validate AI-generated design documents are grounded in local source files by running an automated claim extraction and grep-based back-tracing process. This verifies that proper nouns, state transitions, and numerical values in the artifact actually exist within the correct context of the declared local sources.

Can I verify external citations in AI content against their referenced sources?▼

Yes, you can verify external citations in AI content against their referenced sources using web fetch capabilities. This checks whether fetched external citations actually support the claimed facts, ensuring AI-generated artifacts do not rely on fabricated or unsupported external references.

Does phantom-quench require access to local source files to audit AI artifacts?▼

Yes, phantom-quench requires access to declared local source files to perform grep-based back-tracing of extracted claims. It also requires web fetch capabilities if you need to verify external citations referenced within the AI-generated technical artifacts.

How do I fix structural generation issues when AI creates artifacts without reading source files?▼

To fix structural generation issues when AI creates artifacts without reading source files, you can apply severity classification and process improvement prescriptions. This grades ungrounded claims by risk levels like S, A, or B, and provides specific steps to detect and resolve structural generation gaps.

Related Skills

Explore More Software Engineering Skills→
DTMC-marketplaceDTMC-marketplace

ai-hallucination

Detect and mitigate hallucinations in AI-generated content through factual grounding and source verification.

Official
Intermediate
lovecrushfamilylovecrushfamily
1

vc-autoresearch

Automate iterative gap detection and fixing in software specifications, tests, and UX.

Community
Advanced
hoonsubinhoonsubin

doublecheck

Extract claims from AI-generated text and verify them via web search.

Community
Advanced
kingdonkingdon
4

hallucination-detector

Trace claims to source materials and report grounding confidence.

Community
Basic