Stats
Actions
Tags
From qe-framework
Verifies source credibility and digital content authenticity using the SIFT method. Produces structured verification records with confidence ratings for sources, images, and documents.
How this skill is triggered — by the user, by Claude, or both
Slash command
/qe-framework:Qsource-verifierThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Systematically verifies the credibility of sources and the authenticity of digital content.
Systematically verifies the credibility of sources and the authenticity of digital content.
| Step | Action | Description |
|---|---|---|
| S — Stop | Stop | Do not share or cite before verifying |
| I — Investigate | Investigate the source | Who is behind the information? |
| F — Find | Find other coverage | What do other credible sources say? |
| T — Trace | Trace claims | Find the original source of the claim |
## Source Assessment: [Source Name]
### Basic Identification
- [ ] Real name/organization confirmed
- [ ] Contact information verifiable
- [ ] Professional credentials verifiable
- [ ] Online presence is consistent across platforms
### Expertise Assessment
- [ ] Has relevant expertise for the claim
- [ ] Has track record in the field
- [ ] Recognized by peers
- [ ] No history of spreading misinformation
### Motivation Analysis
- [ ] Potential conflicts of interest identified
- [ ] Financial interest in the outcome?
- [ ] Political/ideological motivation?
- [ ] Related to personal grievances?
### Cross-Verification
- [ ] Claim independently verifiable?
- [ ] Confirmed by other credible sources?
- [ ] Documentary evidence exists?
- [ ] Contradicting sources exist?
| Rating | Criteria |
|---|---|
| High | Real name, relevant expertise, documentary evidence, no conflict of interest |
| Medium | Real name but expertise partially unclear, or indirect interest |
| Low | Anonymous, unverified expertise, potential conflict of interest |
| Suspicious | History of misinformation, clear conflict of interest, unverifiable |
## Image Verification Process
### Step 1: Reverse Image Search
- Google Images (images.google.com)
- TinEye (tineye.com)
- Yandex Images (yandex.com/images) — best for facial recognition
- Bing Visual Search
### Step 2: Metadata (EXIF) Check
- Original capture date/time
- Camera/device information
- GPS coordinates (if present)
- Software used for editing
### Step 3: Image Content Analysis
- Weather conditions (match reported date?)
- Shadows (match time of day?)
- Signs/text (correct language for the location?)
- Architecture (match claimed location?)
### Step 4: Find Original Source
- When did it first appear online?
- Original photographer/source
- Original posting context
- Has it been used in other contexts?
## Account Verification Checklist
### Account History
- Creation date (older = more credible)
- Posting frequency and patterns
- Activity gaps (dormant then suddenly active?)
### Network
- Follower/following ratio
- Follower quality (real accounts vs. bots)
- Connections with verified accounts
### Red Flags
- Recently created account making bold claims
- Sudden changes in topic/tone
- Coordinated behavior with other accounts
- Stock photo profile picture
## Document Verification Steps
### Metadata Inspection
- Creation date and edit history
- Author information
- Software used
- Embedded fonts/images
### Visual Inspection
- Consistent formatting throughout
- Font consistency (no synthesized text)
- Text/image alignment
- Quality consistency across pages
### Content Verification
- Dates internally consistent
- Names spelled consistently throughout
- Reference numbers valid
- Contact information verifiable
- Letterhead matches known examples
### Source Tracing
- How was the document obtained?
- Chain of custody documented?
- Original vs. copy
## Verification Record
**Subject:** [claim/content description]
**Source:** [name/account/platform/URL]
**Verification Date:** [date]
### Verification Steps
#### Step 1: [Description]
- Actions taken:
- Tools/methods used:
- Result:
#### Step 2: [Description]
...
### Cross-Checked Sources
1. [Source 1] — [what it confirmed]
2. [Source 2] — [what it confirmed]
### Contradicting Information
1. [Source] — [contradiction]
### Verdict
- [ ] Verified (true)
- [ ] High confidence (likely true)
- [ ] Unverified (insufficient evidence)
- [ ] Low confidence (contradicting evidence exists)
- [ ] Confirmed false
### Basis for Verdict
[Evidence-based conclusion]
npx claudepluginhub inho-team/qe-framework --plugin qe-frameworkCreates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.