Help us improve
Share bugs, ideas, or general feedback.
From utils
Validates documentation for unsupported claims, made-up metrics, unverifiable statements, comparative hype, and unimplemented features. Generates classified reports with removal or citation recommendations.
npx claudepluginhub jmagly/aiwg --plugin utilsHow this skill is triggered — by the user, by Claude, or both
Slash command
/utils:claims-validatorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Validate documentation for unsupported claims, made-up metrics, and unverifiable statements.
Validates claims through tool execution, avoids superlatives and unsubstantiated metrics. Use when reviewing codebases, analyzing systems, or reporting test results.
Validates factual claims in code reviews, system analysis, documentation, and test reports using tools; prohibits superlatives and unverified metrics.
Verifies factual accuracy of documents against codebase and git history: extracts claims, checks sources, corrects inaccuracies in place, adds summary. Targets reports/plans; auto-detects recent HTML or takes path.
Share bugs, ideas, or general feedback.
Validate documentation for unsupported claims, made-up metrics, and unverifiable statements.
This skill identifies statements that make claims without evidence, including:
When triggered, this skill:
Scans for metric claims:
Identifies unsupported comparatives:
Checks for feature claims:
Validates citations:
Generates report:
# Flagged
"Time Saved: 92-96% (9-15 hours → 45-60 minutes)"
"99x faster routing"
"45x cache speedup"
# Problem
No benchmark data, methodology, or reproducible test
# Fix
Remove claim, or add: "Based on [benchmark/test], measured [how]"
# Flagged
"Budget $20-50/month for moderate use"
"Light usage: ~$10-20/month"
"Enterprise teams may see $100-500+/month"
# Problem
No actual usage data, varies wildly by use case
# Fix
Remove specific numbers, or link to pricing calculator/methodology
# Flagged
"Deploy Full SDLC Framework (2 Minutes)"
"5 minutes, replaces 2-4 hours manual work"
"campaign setup from 2-3 weeks → 1 week"
# Problem
No measurement, varies by project complexity
# Fix
Remove time claims, describe what it does instead
# Flagged
"faster than manual processes"
"more efficient than traditional approaches"
"better than existing solutions"
# Problem
No specific comparison, no baseline defined
# Fix
Remove comparison, or specify exactly what's being compared
# Flagged
"aiwg -migrate-workspace # Optional migration tool"
"Run 'config-validator --fix' to apply automated fixes"
# Problem
Command doesn't exist in codebase
# Fix
Remove until implemented, or mark as "Planned:"
# Flagged
"comprehensive", "revolutionary", "game-changing"
"best-in-class", "industry-leading", "cutting-edge"
"seamless", "effortless", "zero-friction"
# Problem
Subjective claims that can't be verified
# Fix
Replace with specific, factual descriptions
# Claims Validation Report
**Document**: README.md
**Date**: 2025-12-09
**Claims Found**: 12
**Issues**: 8
## Summary
| Category | Found | Unsupported | Action Needed |
|----------|-------|-------------|---------------|
| Metrics | 5 | 4 | Remove or cite |
| Time estimates | 3 | 3 | Remove |
| Cost estimates | 2 | 2 | Remove |
| Comparatives | 1 | 1 | Rephrase |
| Features | 1 | 1 | Remove (not implemented) |
## Issues
### 1. Unsupported Metric
**Line 204**: "Time Savings: 20-98% reduction across 5 core use cases"
**Problem**: No benchmark data or methodology
**Action**: Remove claim
### 2. Cost Estimate Without Data
**Line 1180**: "Light usage: ~$10-20/month"
**Problem**: No actual usage data to support estimate
**Action**: Remove specific numbers
### 3. Non-existent Feature
**Line 583**: "aiwg -migrate-workspace"
**Problem**: Command not implemented
**Action**: Remove until implemented
## Recommendations
1. **Remove all percentage-based time savings claims**
2. **Remove cost estimates** - direct users to measure their own usage
3. **Remove non-existent commands** - only document what exists
4. **Replace superlatives with specifics** - say what it does, not how good it is
User: "Check README.md for unsupported claims"
Output:
"Claims Validation: README.md
Found 8 unsupported claims:
Metrics (4):
- Line 204: '20-98% reduction' - no data
- Line 362: '56-63% time saved' - no data
- Line 588: '99x faster routing' - no benchmark
- Line 589: '45x cache speedup' - no benchmark
Time Estimates (2):
- Line 187: '5 minutes, replaces 2-4 hours' - unmeasured
- Line 366: '2 Minutes' in heading - varies by project
Non-existent Features (2):
- Line 583: 'aiwg -migrate-workspace' - not implemented
- Line 651: 'aiwg -rollback-workspace' - not implemented
Recommendation: Remove these claims. Describe what features
do, not how much time/money they save."
User: "Any BS in this section?"
Output:
"Found 2 issues:
1. 'Revolutionary approach' - marketing speak, be specific
2. '10x productivity gain' - no measurement
Suggest: Replace with factual descriptions of functionality."
This skill complements:
.aiwg/reports/claims-validation.md