cred-1
This MCP server lets AI assistants query the CRED-1 domain credibility dataset.
Check a single domain’s credibility score, category, traffic-light level, and metadata, with automatic www/protocol normalization.
Batch-check up to 100 domains, including
not_foundflags for unlisted domains.Search domains by case-insensitive substring or regex, with configurable result limits (default 20, max 200).
Get dataset statistics: total domains, per-category counts, and version.
List credibility categories with descriptions, typical score ranges, and domain counts.
CRED-1: Open Domain Credibility Dataset
CRED-1 is an open, reproducible domain-level credibility dataset combining multiple openly-licensed source lists with computed enrichment signals. It provides credibility scores for 2,600+ domains (exact count in the latest release) known to publish mis/disinformation, conspiracy theories, or other unreliable content. Community members can report missing domains for inclusion.
🎓 Presented at ACM WebSci 2026 (Braunschweig). Landing page: aloth.github.io/agentic-ai-information-integrity/cred-1.
🤖 MCP-ready: query domain credibility directly from Claude Desktop, Cursor & Windsurf, or any MCP-compatible agent. Jump to setup ↓
Paper: A. Loth, M. Kappes, and M.-O. Pahl, "CRED-1: An Open Multi-Signal Domain Credibility Dataset for Automated Pre-Bunking of Online Misinformation," Preprint, 2026. doi:10.2139/ssrn.6448466
Install
# CLI (global)
npm install -g @aloth/cred1
# Library (project dependency)
npm install @aloth/cred1
# Or try without installing
npx @aloth/cred1 check infowars.com
# Python library, dataset included (pandas optional)
pip install cred1
pip install "cred1[pandas]"import cred1
cred1.check("https://www.infowars.com/article") # dict, or None if not listed
df = cred1.to_dataframe() # whole dataset as a DataFrameRelated MCP server: unphurl-mcp
CLI Usage
# Single domain lookup
cred1 check infowars.com
# 🔴 infowars.com
# Score: 0.073 / 1.000
# Category: conspiracy
# Level: low
# Sources: 2
# Domain not in dataset
cred1 check nytimes.com
# ⚪ nytimes.com
# Not found in CRED-1 dataset: treat as unknown/neutral
# Batch processing (stdin)
echo -e "rt.com\ninfowars.com\nnytimes.com" | cred1 batch
# JSON output
cred1 check breitbart.com --json
# Search
cred1 search "news"
cred1 search "\.ru$"
# Statistics
cred1 stats
cred1 categoriesIn scripts, cred1 check <domain> --exit-code exits 0 when the domain is listed, 1 when it is not and 2 on a usage error. Color follows NO_COLOR and --no-color, and is off when the output is not a terminal. Requires Node 20 or newer.
Domain normalization is automatic: https://www.infowars.com/politics/ resolves to infowars.com.
MCP Server (Claude Desktop / Cursor / Windsurf)
CRED-1 ships an MCP server so AI assistants can check domain credibility directly.
Tools exposed
Tool | Description |
| Check a single domain (score, category, level, metadata) |
| Check up to 100 domains at once |
| Search domains by substring or regex pattern |
| Dataset statistics (total, per-category counts, version) |
| Category taxonomy with descriptions and score ranges |
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"cred1": {
"command": "npx",
"args": ["-y", "@aloth/cred1", "--mcp"]
}
}
}Note:
--mcpis handled by the CLI wrapper, or run the dedicated binary directly.
Alternatively, if the package is installed globally:
{
"mcpServers": {
"cred1": {
"command": "cred1-mcp"
}
}
}Cursor
Add to .cursor/mcp.json (project) or ~/.cursor/mcp.json (global):
{
"mcpServers": {
"cred1": {
"command": "npx",
"args": ["-y", "@aloth/cred1", "--mcp"]
}
}
}Windsurf
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"cred1": {
"command": "npx",
"args": ["-y", "@aloth/cred1", "--mcp"]
}
}
}OpenClaw / any MCP-compatible host
{
"command": "cred1-mcp",
"transport": "stdio"
}Pinning a version
With npx -y @aloth/cred1, npx checks npm for a newer version each time the host starts the server and installs it, and a new version is published every week. A host restart can therefore run a release you have not reviewed. To decide when to update, pin the version in the config:
{
"mcpServers": {
"cred1": {
"command": "npx",
"args": ["-y", "@aloth/cred1@2026.10.27", "--mcp"]
}
}
}Or install a fixed version globally and point the host at the cred1-mcp binary as shown above:
npm install -g @aloth/cred1@2026.10.27Replace 2026.10.27 with the version you reviewed; npm view @aloth/cred1 version prints the latest one.
Most weekly releases only refresh the bundled dataset (data/cred1_compact.json), but every release is built from the current main branch, so code changes ship the same way. To see what changed between two versions, compare their tags, for example v2026-10-20...v2026-10-27. The server reads the dataset from the installed package and makes no network requests. Since 2026.6.14, releases are published from GitHub Actions with npm provenance, which npm audit signatures verifies.
Library Usage
import { checkDomain, searchDomains, getStats } from '@aloth/cred1';
// Single lookup
const result = checkDomain('infowars.com');
// { domain: 'infowars.com', score: 0.073, category: 'conspiracy', level: 'low', sources: 2, domainAge: 27.3, trancoRank: 15889 }
// Not found → null
const unknown = checkDomain('nytimes.com'); // null
// Search by pattern (substring or regex)
const russian = searchDomains('\\.ru$');
// Dataset statistics
const stats = getStats();
// e.g. { totalDomains: 2672, categories: { unreliable: 2001, fake: 233, ... }, version: '2026.11.3' }Traffic-Light Scoring
Level | Score | Emoji | Meaning |
low | ≤ 0.20 | 🔴 | High credibility risk |
mixed | 0.21–0.50 | 🟡 | Unreliable or mixed signals |
ok | > 0.50 | 🟢 | Generally considered reliable |
neutral | not found | ⚪ | Unknown, absence ≠ trustworthy |
Key Features
2,600+ domains with credibility scores (0.0–1.0)
Dual-mode: works as CLI tool and JavaScript library
Fully reproducible: Python pipeline rebuilds the dataset from scratch
Multi-signal scoring combining source labels, domain age, web popularity, fact-check frequency, and threat intelligence
Privacy-preserving: designed for on-device client-side deployment (no server calls needed)
Two openly-licensed sources: no proprietary data dependencies
Domain normalization: handles www., protocols, paths automatically
Dataset Schema
Compact Format (cred1_compact.json)
{
"infowars.com": { "c": "c", "s": 0.073, "n": 2, "d": "1999-10-04", "r": 15889 }
}Field | Description |
| Category code: |
| Credibility score (0.0–1.0, lower = less credible) |
| Number of independent source lists flagging this domain |
| Domain registration date (optional) |
| Tranco Top-1M rank (optional, lower rank = more popular) |
Full Format (cred1_current.json)
{
"infowars.com": {
"category": "fake",
"credibility_score": 0.14,
"domain_age_years": 26.4,
"domain_registered": "1999-10-04T04:00:00Z",
"iffy_factual": "VL",
"iffy_bias": "FN",
"iffy_score": 0.1,
"factcheck_claims": 52,
"safe_browsing_flagged": false,
"score_age": 0.2,
"score_cat": 0.05,
"score_factcheck": 0.0,
"score_iffy": 0.1,
"score_safebrowsing": 0.05,
"score_tranco": 0.1,
"sources": 2,
"tranco_rank": 4382
}
}See CODEBOOK.md for full field documentation.
What the score is, and what it is not
The credibility score is a domain-level signal, not an item-level verdict. It says how a source has been rated by public source-rating initiatives; it does not say whether any individual article is true. Research that uses it as a veracity label is therefore working with a provenance-implied rather than item-verified signal: material drawn from domains scoring below 0.25 can be labelled low-credibility, material from established outlets legitimate, but no single item has been fact-checked. Outlet reputation is a good but imperfect proxy, and that bound should be carried into any reuse.
Rebuilding the Dataset
cd pipeline/
python3 build_dataset.py # Full pipeline
python3 build_dataset.py --step fetch # Download raw data only
python3 build_dataset.py --step merge # Parse + merge (requires prior fetch)
python3 enrich_dataset.py # Add enrichment signals (API keys required)Versioning
CRED-1 uses calendar versioning (CalVer) across all distribution channels:
Channel | Format | Example |
GitHub Release |
| Tag + Zenodo archive |
npm package |
| Same date, dot-separated (valid semver) |
A new version is released weekly with rescored domains. The npm package updates automatically with each GitHub release, no separate version scheme needed.
To pin a specific dataset version:
npm install @aloth/cred1@2026.6.13Production Integrations
Trackless Links: Safari extension for iOS and macOS that checks links locally against CRED-1 (reviewed in Mac & i 4/2026)
RogueGPT: uses CRED-1 to identify low-credibility source domains for controlled fake-news stimulus generation
HuggingFace: dataset mirror for ML pipelines
Contributing
Know a domain that should be in the dataset? Report a missing domain with evidence.
We accept community-reported domains that meet at least one of these criteria:
Listed by Media Bias/Fact Check (MBFC)
Flagged by an IFCN-certified fact-checker
Subject to press council or regulatory action
Documented in Wikipedia with credibility concerns
Referenced in published academic research
Accepted domains are added to the pipeline and appear in the next weekly dataset release.
Citation
If you use CRED-1 in your work, please cite both the paper and the dataset version you used. The paper describes the method; the Zenodo record pins the exact data, which is what makes your results reproducible.
Paper:
@misc{loth2026cred1,
author = {Loth, Alexander and Kappes, Martin and Pahl, Marc-Oliver},
title = {{CRED-1}: An Open Multi-Signal Domain Credibility Dataset for Automated Pre-Bunking of Online Misinformation},
year = {2026},
howpublished = {Preprint},
doi = {10.2139/ssrn.6448466},
url = {https://ssrn.com/abstract=6448466},
urldate = {2026-08-25}
}Dataset:
@dataset{loth2026cred1data,
author = {Loth, Alexander},
title = {{CRED-1}: An Open Multi-Signal Domain Credibility Dataset},
year = {2026},
publisher = {Zenodo},
version = {v2026-03-17},
doi = {10.5281/zenodo.18769459},
url = {https://doi.org/10.5281/zenodo.18769459}
}The DOI above always resolves to the latest release. To cite a specific weekly snapshot, use the version DOI shown on that release's Zenodo page.
Related Projects
CRED-1 is one strand of a wider research program on generated disinformation. The domain scores are used to select low-credibility sources for stimulus generation in RogueGPT:
Project | Description |
Empirical platform for evaluating AI-generated news authenticity | |
Controlled stimulus generator; draws fake-news sources from CRED-1 scores | |
Expert survey on GenAI disinformation threats and countermeasures | |
iOS app for C2PA content credentials and EXIF verification | |
Reproducibility bundle for a benchmark audit of AI-text detection |
Companion data deposits
CRED-1 is archived alongside three companion datasets, which are versioned together so that they can be cited as one coherent set:
Deposit | Concept DOI | Current version |
RogueGPT Stimulus Corpus | 22225536 (v1.2.0) | |
JudgeGPT Human Perception Data | 22226580 (v1.2.0) | |
Expert Survey | 22226924 (v2.1.0) |
Related Publications
Interrupting the Chain: Human Perception of AI-Generated Disinformation Through a Kill Chain Lens (INFORMATIK 2026, LNI). Positions domain-level credibility as a Delivery-stage signal. arXiv:2608.21389
Blessing or Curse? A Survey on the Impact of Generative AI on Fake News (2024). arXiv:2404.03021
License
The dataset is everything in data/ and the releases built from it, including the copies on Zenodo and Hugging Face; LICENSE holds the CC BY 4.0 text. The code is src/, pipeline/ and examples/; LICENSE-MIT holds the MIT text. The npm package @aloth/cred1 ships both, so its license field reads MIT AND CC-BY-4.0.
The dataset is an aggregation of domain-level ratings from public source-rating initiatives. The aggregation, scoring and documentation are this project's own work and are released under CC BY 4.0; the underlying ratings remain attributable to their respective initiatives, which are named per record and listed in ATTRIBUTION.md. No access restriction applies to the dataset.
Author
Alexander Loth · alexloth.com · @xlth · ORCID
Available Tools
5 toolsbatch_checkA
Check multiple domains at once (max 100). Returns an array of results with the same fields as check_domain. Domains not found in the dataset are included with a "not_found" flag.
Inputs: domains (required).
check_domainA
Check a single domain's credibility score and category using the CRED-1 dataset. Returns score (0.0–1.0, lower = less credible), category, traffic-light level, and metadata.
Inputs: domain (required).
get_categoriesA
List all credibility categories with their descriptions, typical score ranges, and domain counts from the dataset.
Inputs: no named parameters.
get_statsA
Get summary statistics for the CRED-1 dataset: total domain count, per-category breakdown, and dataset version.
Inputs: no named parameters.
search_domainsA
Search for domains matching a substring or regex pattern. Useful for discovering all entries related to a topic or checking variations of a domain name.
Inputs: pattern (required), limit.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 5 tools
Each tool has a clearly distinct purpose: single lookup, bulk lookup, pattern search, dataset-level stats, and category metadata. There is no realistic scenario where an agent would confuse check_domain with search_domains or get_stats, and batch_check is explicitly distinguished by its multi-domain input and result array.
Four tools follow a clean verb_noun pattern (check_domain, search_domains, get_stats, get_categories). batch_check deviates slightly by putting the noun first, but it remains readable and the intent is obvious, so this is a minor deviation rather than an inconsistency problem.
Five tools is a tight, well-scoped surface for a domain credibility dataset: one tool per natural operation (read one, read many, search, stats, taxonomy). Nothing appears redundant or missing at the count level.
For a read-only dataset server, the surface covers all expected workflows: single lookup, bulk lookup with not_found handling, discovery via substring/regex search, aggregate statistics, and category taxonomy. There are no dead ends an agent would hit in normal use.
Maintenance
Related MCP Connectors
Verify claims with verdict, confidence & cited sources; batch verify, source checks, daily brief.
Is this website a scam? Dated, sourced fraud assessments of domains. Free, read-only, no key.
Check whether a website is trustworthy: trust score, verdict, red flags and a citable report.
Domain data: Top 1M domain ranks, compromised checks, new domain search, registered domain lists.
Related MCP Servers
- AlicenseAqualityBmaintenanceSearch 1,500+ pre-verified, cited knowledge units across 16 domains. 6 tools: query, batch query, get unit, list domains, suggest topics, report issues. Free, no API key required.6237 npm1MIT
- AlicenseAqualityDmaintenanceURL intelligence for AI agents. One URL in, structured security and data quality signals out across 7 dimensions. 13 tools, risk score 0-100 with 23 configurable weights.1655 npm1MIT
- FlicenseNot gradedqualityFmaintenancePay-per-call API that verifies whether a domain belongs to a real business. Returns a verdict (real/likely_real/uncertain/likely_fake/fake), a 0-100 score, and signals (WHOIS via RDAP, SSL via Certificate Transparency, homepage LLM judgment, contacts, social) — for KYB, vendor screening, fraud checks, and lead qualification.-
- AlicenseNot gradedqualityCmaintenanceProvides AI agents with zero-cost, zero-dependency trust evaluation for domains, URLs, wallets, APIs, and IPs, returning a 0-100 score with SSL, DNS, WHOIS, security header, and content sub-scores, plus batch comparison to rank multiple targets.MIT