Scoring Methodology

How MCP Rated evaluates every MCP server through published criteria, evidence summaries, and coverage notes. Version 2.1 · Last updated March 2026.

Philosophy

  • →Independent by design. Scores are based on published criteria rather than vendor claims or paid placement.
  • →Evidence before opinion. Protocol behavior, repository signals, documentation quality, and server metadata shape the review.
  • →Coverage is never hidden. When an area cannot be assessed fairly, the limitation is disclosed alongside the score.
  • →Versioned methodology. Scoring changes are tracked so historical comparisons have context.

Score Taxonomy

Each server is reviewed across six dimensions. The overall score translates those dimensions into a 0–100 view for quick comparison.

25%

Reliability

Protocol conformance, connection stability, schema validity, error handling

25%

Security

Instruction-safety review, dependency audit, credential exposure checks, authentication

15%

Setup

Ease of getting started: README, setup guides, transport DX

15%

Documentation

Quality & completeness of descriptions, schemas, categories

10%

Compatibility

Transport support, schema completeness, tool integration depth

10%

Maintenance

GitHub health signals, adjusted for project scale

Coverage-Aware Scoring

Not every server exposes the same surface area. If a dimension cannot be assessed fairly, MCP Rated explains the limitation and avoids presenting a coverage gap as a failure.

Reliability Status Classification

Some servers require OAuth, a vendor sandbox, or a specific transport environment. Instead of treating those constraints as failures, reliability is shown with one of three coverage statuses:

measured

Full protocol test completed — score reflects actual connection stability, schema validity, and error handling.

partial

Connected to the server but could not test individual tools. Score shown with caveat.

not_testable

Could not complete enough testing for a reliability score due to access, transport, or sandbox limitations. Shown as N/A — not a negative signal.

When reliability cannot be measured, the detail page presents that limitation directly rather than turning it into a negative reliability claim.

Evidence-Based Review

MCP Rated favors evidence that can be gathered without changing third-party systems. Destructive or state-changing behavior is considered only when a suitable sandbox or vendor-provided environment exists.

  • →Measured claims. Public conclusions are tied to evidence that can be reviewed and explained.
  • →Summarized evidence. Detail pages describe relevant outcomes and caveats without exposing sensitive internal material.
  • →Consistent interpretation. Similar findings are grouped into the same review categories across servers.

Security Scoring

Security review combines public vulnerability data, repository signals, credential exposure indicators, and unsafe-instruction review. Supplemental editorial triage may inform prioritization, but not the sole basis for a negative public score.

Instruction safety

Unsafe-instruction and prompt-injection review

Dependencies

Lockfile presence, known vulnerabilities via OSV and GitHub Advisory

Secrets

Credential exposure and secret hygiene signals

Auth

Authentication method appropriateness for the transport type

Evidence Sources

Vulnerability advisories

Known public CVE and GHSA records

public data

Repository security signals

Dependency and maintenance indicators

repo signal

Credential exposure indicators

Signals that suggest unsafe secret handling

review signal

Instruction-safety review

Signals that suggest unsafe tool guidance

review signal

Supplemental triage

Additional context considered during review

editorial

Quality Gate

Server pages are indexed only after meeting quality checks for coverage, data completeness, and scoring consistency. Pages that don't meet the quality bar remain hidden from search engines until sufficient evidence is available.

Badge Criteria

Badges are awarded based on score thresholds across reliability, security, and overall quality.

Lab Tested

Server has been tested with sufficient coverage and meets minimum quality standards.

Vendor Verified

Server demonstrates high reliability and security scores from structured testing.

Security Scanned

Server has passed security scanning with satisfactory results.

Vulnerability Disclosure Policy

We publicly list only already-disclosed vulnerabilities from authoritative databases (OSV, GitHub Advisory). Newly discovered findings from internal scans are kept private until responsibly disclosed to the vendor.

public_known

Known public CVE/GHSA/advisory — listed immediately

private_triage

New finding under investigation — not shown publicly

private_disclosed

Confirmed and reported to vendor, awaiting fix

public_disclosed

Disclosure process complete — listed publicly

Transparency & Re-test Policy

  • →Summarized evidence and review caveats are visible on server detail pages.
  • →Maintainers can request a re-test by contacting us.
  • →Methodology changes are tracked and versioned.
  • →Paid features and coverage limitations are disclosed separately so scores do not overstate what was actually tested.