Is novada-proxy-9c53c0 safe to install?

SaferSkills independently audited novada-proxy-9c53c0 (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.

CategoryWeightCategory scoreContribution

Securityprompt, exec, net, exfil, eval

35%

100

35.0 pts

Supply chainhash, typosquat, maintainer, lockfile

20%

100

20.0 pts

Maintenancestaleness, pinning, CI

15%

100

15.0 pts

TransparencySKILL.md, perms, README

15%

100

15.0 pts

Communityinstalls, verify, response

15%

100

15.0 pts

Findings & checks · 0 flagged

Securityscore 100 · 0 findings

✓—All security checks passedNo findings in this category for the latest scan.pass

Supply chainscore 100 · 0 findings

✓—All supply chain checks passedNo findings in this category for the latest scan.pass

Maintenancescore 100 · 0 findings

✓—All maintenance checks passedNo findings in this category for the latest scan.pass

Transparencyscore 100 · 0 findings

✓—All transparency checks passedNo findings in this category for the latest scan.pass

Communityscore 100 · 0 findings

✓—All community checks passedNo findings in this category for the latest scan.pass

The primary manifest — the file an agent reads to learn what this artifact does.

skill/SKILL.md · 1 file

skill/SKILL.md6.6 KB · Markdown

name: novada-proxy

description: >-

Residential proxy tools for AI agents — fetch, crawl, search, extract,

render through 2M+ IPs. Bypass anti-bot, geo-target 195+ countries.

Use when: scraping websites, researching topics, extracting structured

data, or any task that needs web access through a proxy.

version: 1.8.2

author: NovadaLabs

platform: clawhub

install:

mcp:

command: npx

args: ["-y", "novada-proxy-mcp"]

transport: stdio

env:

NOVADA_PROXY_USER: "required — proxy username"

NOVADA_PROXY_PASS: "required — proxy password"

NOVADA_API_KEY: "optional — enables novada_proxy_search"

NOVADA_BROWSER_WS: "optional — enables novada_proxy_render"

security:

network: outbound only (proxy requests to target URLs)

credentials: proxy credentials via env vars

filesystem: none

telemetry: none

tags: [proxy, scraping, web-access, anti-bot, geo-targeting, mcp, residential-ip]

trigger:

- "scrape"

- "fetch through proxy"

- "bypass cloudflare"

- "geo-target"

- "residential proxy"

- "extract product data"

- "crawl site"

- "web search"

- "research"

- "competitive analysis"

- "topic investigation"

Novada Proxy — Skill for Claude Code

When to Use This Skill

Fetching URLs that block direct requests (Cloudflare, Akamai, DataDome)
Scraping e-commerce, news, or professional sites (Amazon, LinkedIn, etc.)
Extracting structured fields (title, price, rating) without HTML parsing
Researching topics (search + batch read pipeline)
Crawling entire sites (discover URLs + parallel fetch)
Accessing geo-restricted content from 195+ countries
Multi-step flows requiring the same IP (login, pagination)
Rendering JS-heavy SPAs that return blank with standard HTTP

Setup

# Core (proxy fetch)
claude mcp add novada-proxy-mcp \
  -e NOVADA_PROXY_USER=your_username \
  -e NOVADA_PROXY_PASS=your_password \
  -- npx -y novada-proxy-mcp

# All tools (proxy + search + browser render)
claude mcp add novada-proxy-mcp \
  -e NOVADA_PROXY_USER=your_username \
  -e NOVADA_PROXY_PASS=your_password \
  -e NOVADA_API_KEY=your_key \
  -e NOVADA_BROWSER_WS=your_wss_url \
  -- npx -y novada-proxy-mcp

Credentials: novada.com -> Dashboard -> Residential Proxies

Tool Decision Tree

Need to fetch a URL?
  Static HTML page             -> novada_proxy_fetch
  Need specific fields         -> novada_proxy_extract
  JS-heavy / SPA / blank page  -> novada_proxy_render (5 credits)

Need multiple URLs?
  Have the URLs already        -> novada_proxy_batch_fetch (2-20 URLs, up to 5x parallel)
  Need links from ONE page     -> novada_proxy_map -> novada_proxy_batch_fetch
  Need to crawl a whole site   -> novada_proxy_crawl (recursive BFS, depth 1-5)

Need to find URLs by topic?    -> novada_proxy_search -> novada_proxy_batch_fetch

Need research on a topic?      -> novada_proxy_research (search + read + summarize)

Need same IP across calls?     -> novada_proxy_session (sticky IP via session_id)

Need to check connectivity?    -> novada_proxy_status

Tools Reference

Tool	Purpose	Credits	Key Params
`novada_proxy_fetch`	Fetch single URL via residential proxy	1	`url`, `country`, `format`, `session_id`
`novada_proxy_batch_fetch`	Fetch 2-20 URLs concurrently	1/URL	`urls`, `concurrency` (1-5), `country`
`novada_proxy_extract`	Extract structured fields from URL	1 (5 if render fallback)	`url`, `fields[]`, `render_fallback`
`novada_proxy_map`	Discover all internal links on a page	1	`url`, `limit` (10-200)
`novada_proxy_crawl`	Recursive site crawl (BFS, depth 1-5)	1/page	`url`, `depth` (1-5), `limit`, `include_content`
`novada_proxy_session`	Sticky session (same IP across calls)	1 (3 if verify)	`session_id`, `url`, `verify_sticky`
`novada_proxy_search`	Google search as structured JSON	1	`query`, `num` (1-20), `country`
`novada_proxy_render`	Render JS pages via real Chromium [BETA]	5	`url`, `format`, `wait_for` (CSS selector)
`novada_proxy_research`	Search + read + summarize on a topic	varies	`query`, `num_sources` (1-10), `country`
`novada_proxy_status`	Check proxy health + version	1	_(none)_

Common Workflows

1. Site Crawl (map -> batch)

novada_proxy_map(url, limit=50)           -> internal_urls[]
novada_proxy_batch_fetch(urls, concurrency=5) -> all pages in parallel

2. Research (search -> batch)

novada_proxy_search(query, num=5)         -> titles + URLs + snippets
novada_proxy_batch_fetch(urls, format="markdown") -> full page content

3. Sticky Session (login + multi-page)

novada_proxy_session(session_id="job001", url="/login")
novada_proxy_session(session_id="job001", url="/dashboard")
novada_proxy_session(session_id="job001", url="/data/page/1")

4. Price Monitoring (geo comparison)

novada_proxy_fetch(url=product_url, country="US")
novada_proxy_fetch(url=product_url, country="DE")
novada_proxy_fetch(url=product_url, country="JP")
# 2nd call per URL+country is a cache hit (0 credits)

5. Structured Extraction (with JS fallback)

novada_proxy_extract(url, fields=["title","price","rating"], render_fallback=true)
# Auto-escalates to Chromium if proxy fetch is blocked

Error Recovery

Error Code	Recoverable	Next Step
`BOT_DETECTION_SUSPECTED`	Yes	Retry with `novada_proxy_render`, or different `country`
`TLS_ERROR`	Yes	Retry with different `country`, or use `novada_proxy_render`
`TIMEOUT`	Yes	Increase `timeout` param (max 120s), or retry
`RATE_LIMITED`	Yes	Wait 5s, then retry
`NETWORK_ERROR`	No	DNS failed. Check URL/domain.
`SESSION_STICKINESS_FAILED`	Yes	Retry with `verify_sticky: true`
`INVALID_INPUT`	No	Fix params per tool's inputSchema
`PROVIDER_NOT_CONFIGURED`	No	Set env vars and restart MCP

Escalation chain: novada_proxy_fetch fails -> try different country -> try novada_proxy_render -> check novada_proxy_status

Cost Guide

Tool	Credits	Cache?
fetch	1	Yes (300s TTL)
batch_fetch	1 per URL	Yes
extract	1 (5 if render fallback)	No
map	1	No
crawl	1 per page crawled	Yes (pages use fetch cache)
session	1 (3 if verify_sticky)	Never (sticky routing)
search	1	No
render	5	No
status	1	No

Optimization: Always try fetch before render (5x cheaper). Cache key = url + format + country. Repeated calls within 300s = 0 credits (meta.cache_hit: true). Check meta.quota.credits_estimated on every response.

novada-proxy-9c53c0— agent skill

Is novada-proxy-9c53c0 safe to install?

Novada Proxy — Skill for Claude Code

When to Use This Skill

Setup

Tool Decision Tree

Tools Reference

Common Workflows

Error Recovery

Cost Guide

Audit the pieces. Scan the whole. Decide.