# io.github.NovadaLabs/proxy4agents-mcp (npm · bestproxy4agents-mcp)

Residential proxy MCP for AI agents — geo-targeting, sticky sessions, anti-bot bypass.

- Trust score: 63/100 (medium)
- Change this week: +24
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-08-03

## Components

- npm · `bestproxy4agents-mcp`: 63/100 (this document), [markdown](https://verifymcp.io/servers/novadalabs-proxy4agents-mcp/bestproxy4agents-mcp.md), [page](https://verifymcp.io/servers/novadalabs-proxy4agents-mcp/bestproxy4agents-mcp)

## Channel facts

- Registry: `npm`
- Package: `bestproxy4agents-mcp`
- Version: `1.7.0`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-08-03.

- **Supply Chain Security**: 87/100
  - No malware found by supply-chain analysis.
  - Only part of the dependency tree could be resolved (189 of 190), so this covers what we could see, not the whole tree.
  - No install/post-install scripts declared.
  - Only part of the dependency tree could be resolved (189 of 190), so this covers what we could see, not the whole tree.
- **Provenance & Transparency**: 19/100
  - Repository check failed: the declared repository URL redirects; it must resolve directly.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 111 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 69/100
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 866 tokens (~144/item across 6 items; 6 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 27/100
  - Stability observed for 8 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 99/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 96% of tool parameters carry a description.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### Claude

```bash
claude mcp add novadalabs-proxy4agents-mcp -- npx -y bestproxy4agents-mcp
```

### Codex

```bash
codex mcp add novadalabs-proxy4agents-mcp -- npx -y bestproxy4agents-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "novadalabs-proxy4agents-mcp": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "bestproxy4agents-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add novadalabs-proxy4agents-mcp --command npx --arg -y --arg bestproxy4agents-mcp
```

### Hermes

```yaml
mcp_servers:
  novadalabs-proxy4agents-mcp:
    command: "npx"
    args: ["-y", "bestproxy4agents-mcp"]
```

### Other

```json
{
  "mcpServers": {
    "novadalabs-proxy4agents-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "bestproxy4agents-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-08-03 (score 63, +33)

- [security regression] Provenance: unverified → fail
- [security improvement] Install scripts: unverified → pass
- [security improvement] Known CVEs: unverified → partial
- [functional improvement] Schema quality: unverified → good
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Stability: unverified → 0.27
- [functional improvement] Maintenance: unverified → pass
- [functional improvement] Dependency health: unverified → partial
- [functional improvement] License: unverified → pass
- [functional] Licence: MIT

### 2026-08-02 (score 30, +13)

- [functional regression] Dependency health: partial → unverified
- [functional improvement] Tool coverage: unverified → 100

### 2026-08-01 (score 17, −13)

- [functional regression] Tool coverage: 100 → unverified
- [functional improvement] Dependency health: unverified → partial

### 2026-07-31 (score 30, −9)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-07-30 (score 39, +21)

- [functional improvement] Tool coverage: unverified → 100

### 2026-07-28 (score 18, −21)

- [functional regression] Tool coverage: 100 → unverified
- [functional] First check of Schema quality: unverified

### 2026-07-27 (score 39)

First indexed and scored.

## MCP tools (6)

### `agentproxy_fetch` (~188 tokens)

Fetch any URL through a residential proxy network (2M+ IPs, 195 countries, anti-bot bypass). Works on Amazon, LinkedIn, Cloudflare-protected pages, and most commercial sites. Use agentproxy_render instead for JavaScript-heavy pages that need a real browser.

Input parameters:

- `city` (string): City-level targeting: newyork, london, tokyo, ...
- `country` (string): 2-letter country code: US, DE, JP, GB, BR, ... (195+ options)
- `format` (string): markdown strips HTML tags; raw returns full HTML
- `session_id` (string): Use the same ID across calls to stay on the same IP (letters, numbers, underscores only — no hyphens)
- `timeout` (number): Timeout in seconds (1-120)
- `url` (string, required): The URL to fetch

### `agentproxy_render` (~142 tokens)

[BETA] Render a JavaScript-heavy page using Novada's Browser API (real Chromium, full JS execution). Use this for SPAs, React/Vue apps, and sites that require JavaScript to load content. Requires NOVADA_BROWSER_WS env var. For static/HTML pages, agentproxy_fetch is faster.

Input parameters:

- `format` (string): markdown strips tags; html returns full HTML; text strips tags + links
- `timeout` (number): Timeout in seconds (5-120)
- `url` (string, required): The URL to render
- `wait_for` (string): CSS selector to wait for before extracting (e.g. '.product-title')

### `agentproxy_search` (~124 tokens)

Structured web search via Novada (Google). Returns titles, URLs, and descriptions — no HTML parsing needed. Best for finding pages and factual queries. For reading a specific URL, use agentproxy_fetch instead.

Input parameters:

- `country` (string): Country for localized results (e.g. us, uk, de)
- `engine` (string): Search engine (google)
- `language` (string): Language code (e.g. en, zh, de)
- `num` (number): Number of results (1-20)
- `query` (string, required): The search query

### `agentproxy_extract` (~219 tokens)

Extract structured data from any URL through a residential proxy. Fetches the page and extracts named fields (title, price, description, rating, author, date, image, links, headings, etc.) using meta tags, Open Graph, JSON-LD, and HTML patterns. Returns clean key-value output — no HTML parsing needed by the agent.

Input parameters:

- `city` (string): City-level targeting (e.g. newyork, london, tokyo)
- `country` (string): 2-letter country code for geo-targeting (e.g. US, DE, JP)
- `fields` (array, required): Field names to extract: title, price, currency, description, image, rating, review_count, author, date, url, links, headings, h2, or any JSON-LD/meta field name
- `session_id` (string): Reuse same ID to keep the same IP across calls
- `timeout` (number): Timeout in seconds (1-120)
- `url` (string, required): The URL to extract data from

### `agentproxy_session` (~152 tokens)

Fetch a URL with a sticky session — the same residential IP is used for every call with the same session_id. Use this for multi-step workflows where IP consistency matters: login flows, paginated scraping, price monitoring.

Input parameters:

- `city` (string): City-level targeting (e.g. newyork, london, tokyo)
- `country` (string): 2-letter country code
- `format` (string)
- `session_id` (string, required): Unique session ID — reuse the same value to keep the same IP (letters, numbers, underscores only — no hyphens)
- `timeout` (number): Timeout in seconds (1-120)
- `url` (string, required): The URL to fetch

### `agentproxy_status` (~41 tokens)

Check proxy network health — live node count, device types, active provider, service status. Use before starting a large scraping workflow to confirm the network is healthy.

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/novadalabs-proxy4agents-mcp/bestproxy4agents-mcp#diagnostics

## Score history

- 2026-08-03: 63
- 2026-08-02: 30
- 2026-08-01: 17
- 2026-07-31: 30
- 2026-07-30: 39
- 2026-07-28: 18
- 2026-07-27: 39

## Links

- npm package: https://www.npmjs.com/package/bestproxy4agents-mcp
- Socket report: https://socket.dev/npm/package/bestproxy4agents-mcp
- Changelog RSS feed: https://verifymcp.io/servers/novadalabs-proxy4agents-mcp/bestproxy4agents-mcp/changelog.xml
- Changelog JSON feed: https://verifymcp.io/servers/novadalabs-proxy4agents-mcp/bestproxy4agents-mcp/changelog.json
- HTML version of this page: https://verifymcp.io/servers/novadalabs-proxy4agents-mcp/bestproxy4agents-mcp