
Agentsvc MCP server
Which MCP servers are Agentsvc's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-10-26
- Registry nameio.agentsvc/servicesverified in the MCP registry
The registry name sits under agentsvc.io. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-10-26
- Tools12
- Server calls itselfagentsvc
- Server version2.6.0
- Protocol version2025-06-18
On 2026-10-26 the server answered initialize and tools/list without a sign-in. First answer in 165 ms.
- Its tools do convert web pages into markdown for an LLM, but not screen companies for sanctions and validate VAT numbers.
- Checked on 2 of the 7 days it takes to judge reliability.
Once connected
- For "convert web pages into markdown for an LLM", the likely first call is web_scrape.
- Parameters are untyped or undescribed in 4 of 12 tools: check arguments before calling.
Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.
- INITE MCP scoreafter a weeka score needs 7 daily checks: 2 so far
- What an agent can do with it1 of 2 use cases
- Answeredmeasuring2 of the 7 daily checks it takes to print a share
- Median first answer509 ms
- Space in an agent's context≈ 3,582 tokensthe definitions of all its tools (12), which an agent carries on every turn
- Tool definitionsA · 94 of 100
- Typed results0 of 12 toolsdeclared outputSchema: results come typed, read once
- Parameters without a description: in 4 of 12
What an agent can do with it
- can do · convert web pages into markdown for an LLM
- cannot · screen companies for sanctions and validate VAT numbers
- not sure · watch web pages and wallets for changes
Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?
Worth a look
None of the five warning signs in these definitions.
A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →
find_serviceFree. Search all 127 agentsvc.io services by task (e.g. "transcribe audio", "company register", "token rug pull", "pdf to text"). Returns the best matches with price, input schema and a free example input. Then run one with call_service.
call_serviceRun any agentsvc.io service by name with its input, e.g. {"service": "pdf-extract", "input": {"url": "https://..."}}. Use find_service first to get the name and input schema. Costs $0.002 to $0.6 USDC depending on the service (shown by…
ip_lookupIP lookup and geolocation: resolve any public IPv4 or IPv6 address to location and network data from a local database (no third-party call, answers in milliseconds): country, country_code, region, city, lat/lon, IANA timezone, continent…
translateTranslate text between 21 languages: en, de, fr, es, it, pt, nl, pl, sv, cs, el, ru, uk, tr, ar, hi, id, vi, zh, ja, ko. Runs on our own server (Argos Translate), so the text is not sent to a third party. Max 1000 characters per call…
news_searchNews search: latest news articles and headlines by keyword with title, url, source and published_at from our own index of about 100,000 English web articles a day (built from the GDELT Project, thousands of outlets, last 7 days, updated…
web_scrapeWeb page to markdown: read any web page (URL) as clean, LLM-ready Markdown, the web scraper for agents. Default tool for reading ONE web page whose URL you have: returns clean Markdown (title, meta description, main content with headings…
web_scrape_batchBulk web scraper: read up to 20 web pages in ONE paid call, each as clean, LLM-ready Markdown with title and meta description. One payment instead of 20, pages fetched in parallel, duplicate URLs read once. Use when your agent works…
sec_filingsUS public company data from SEC EDGAR: company profile, recent filings (10-K, 10-Q, 8-K, Form 4 insider trades, S-1 ...) with direct document links, and optionally the latest annual financials (revenue, net income, assets, liabilities…
email_verifyEmail verification: checks whether an email address really exists by asking its mail server (SMTP RCPT TO probe; no email is sent), on top of syntax, MX, disposable and role checks. Returns status deliverable, undeliverable (mailbox does…
satellite_imageDaily true-color satellite image of any area from NASA GIBS (MODIS Terra/Aqua or VIIRS, about 250 m per pixel, coastlines overlaid) as base64 JPEG. Useful to see clouds, smoke, snow, floods or fires on a given day. Params: place ("Zurich…
token_verdictInstant buy-or-avoid verdict for a token, with a shareable result page and preview card (share_url): send a contract address (Base, Ethereum, BSC, Arbitrum, Polygon, Avalanche, Optimism) or a Solana mint, get a grade A to F, a label…
audio_transcribeAudio transcription and speech to text: transcribe a voice note, meeting, podcast or call recording into text with timestamps, in about 100 languages with automatic language detection. Runs OpenAI Whisper (small by default, base for…
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"services": {
"type": "http",
"url": "https://agentsvc.io/mcp"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http agentsvc https://agentsvc.io/mcpclaude.ai, ChatGPT and other clients: add https://agentsvc.io/mcp as a custom connector in the settings.
OpenAPI spec
133 paths, read 2026-10-26. Open the spec ↗
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Cookies
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.