@portkey-ai/mcp-tool-filter

Name: @portkey-ai/mcp-tool-filter
Rating: 3.51 (1 reviews)
Author: jumbld

#293 · by jumbld

llm reasoning tool-use vision 200k context

3.51/ 5.00

trustedBeta

Mar 5, 2026 at 8:00 AM6 signals analysedNo manual reviews · fully automated

Trust Signal Breakdown

high23 sub-signals across 6 dimensions

CVEs, dependency health, and supply chain integrity

3 of 3 sub-signals with data

Known CVEs40%5.0

No known CVEs

via OSV.dev

Dependency Health30%5.0

2 dependencies (minimal)

via npm / PyPI

Supply Chain30%5.0

Supply chain analyzed, no transitive CVEs

via npm provenance

Uptime, latency, error rates, and incident history

4 of 4 sub-signals with data

Uptime35%5.0

100.00% over 11 checks

via Health checks

Response Latency25%3.0

p99: 688ms, p50: 390ms

via Health checks

Error Rate20%5.0

0.00% error rate (0/11)

via Health checks

Incident History20%3.0

2 incidents in last 90 days

via Incidents table

Commit recency, release cadence, issue response, CI/CD

3 of 4 sub-signals with data

Commit Recency37%2.0

via GitHub

Release Cadence31%2.0

via GitHub

Issue Responseno data—

Weight redistributed to sub-signals with data

CI/CD Presence31%2.0

via GitHub Actions

Downloads, stars, dependents, and growth trajectory

3 of 4 sub-signals with data

Download Volume43%1.0

13 weekly downloads

via npm / PyPI

GitHub Stars36%1.0

36 stars

via GitHub

Dependent Packagesno data—

Weight redistributed to sub-signals with data

Growth Trend21%1.0

-23.5% week-over-week

via npm

License, documentation, security policy, changelog

4 of 4 sub-signals with data

Open Source30%5.0

Public repo with OSI-approved license (mit)

via GitHub

Documentation25%5.0

Docs site present with comprehensive README (>2000 bytes + examples)

via GitHub

Security Policy20%2.0

No SECURITY.md found

via GitHub

Changelog25%5.0

CHANGELOG.md present and releases exist

via GitHub

Track record, org maturity, community standing

4 of 4 sub-signals with data

Track Record30%3.5

Internal: 1.0 (0 services), External: 3.5 (352 followers, 11123 stars)

via Fabric index

Org Maturity30%4.5

Organization, 2.9 years old

via GitHub

Community Standing20%3.0

34 public repositories

via GitHub

Cross-Platform20%3.0

Present on 2 platform(s): github, npm

via Registry scan

About this score

Scored across 23 sub-signals in 6 dimensionsScoring engine v1 (beta) — actively being expandedPhase 1: Core sub-signal architecture (live)Phase 2: Permission scope & expanded collection (in progress)

Trust AssessmentAI Assessment

@portkey-ai/mcp-tool-filter is an MIT-licensed npm package by jumbld that provides semantic tool filtering for MCP servers using embedding similarity, with dependencies on @xenova/transformers and openai. The package shows critical maintenance concerns with a 0.00/5.00 maintenance score and minimal adoption at 16 weekly downloads, indicating an early-stage or potentially abandoned project. While no vulnerabilities are currently detected, the lack of maintenance activity and transparency metrics (0.00/5.00) makes this package unsuitable for production use without direct maintainer engagement.

Generated by Fabric AI · Mar 4, 2026 at 4:57 AM

Package Availability (30d)

100.00%

p50: 390ms · p99: 688ms

Avg Latency

342ms

averaged across 30d health checks

Weekly Downloads

13-24%

npm weekly

Transparency & Compliance4/6 passed

✓