# SwogeAgentic (remote · web-production-220a3.up.railway.app)

Crypto intel for AI agents: token scans, SWOGE AI colony data, OSINT, web search, images, video

- Trust score: 71/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-28

## Components

- remote · `web-production-220a3.up.railway.app`: 71/100 (this document), [markdown](https://verifymcp.io/servers/dog-swoleeswoge-swogeagentic/web-production-220a3.md), [page](https://verifymcp.io/servers/dog-swoleeswoge-swogeagentic/web-production-220a3)

## Channel facts

- Endpoint: `https://web-production-220a3.up.railway.app/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.0.1`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, 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 2026-09-28.

- **Endpoint Security**: 74/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one.
  - HTTPS is enforced; there's no plaintext access path.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 70/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 3029 tokens (~168/item across 18 items; 18 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 7/100
  - Stability observed for 2 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 18 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 19 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

## Install

### How do I install the SwogeAgentic MCP server?

SwogeAgentic is a hosted endpoint at https://web-production-220a3.up.railway.app/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http dog-swoleeswoge-swogeagentic 'https://web-production-220a3.up.railway.app/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "dog-swoleeswoge-swogeagentic": {
      "url": "https://web-production-220a3.up.railway.app/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "dog-swoleeswoge-swogeagentic": {
      "type": "http",
      "url": "https://web-production-220a3.up.railway.app/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.dog-swoleeswoge-swogeagentic]
url = "https://web-production-220a3.up.railway.app/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "dog-swoleeswoge-swogeagentic": {
      "type": "remote",
      "url": "https://web-production-220a3.up.railway.app/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add dog-swoleeswoge-swogeagentic --url 'https://web-production-220a3.up.railway.app/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  dog-swoleeswoge-swogeagentic:
    url: "https://web-production-220a3.up.railway.app/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "dog-swoleeswoge-swogeagentic": {
      "Transport": "http",
      "Url": "https://web-production-220a3.up.railway.app/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add dog-swoleeswoge-swogeagentic -t streamable-http -u 'https://web-production-220a3.up.railway.app/mcp'
```

### Other

```json
{
  "mcpServers": {
    "dog-swoleeswoge-swogeagentic": {
      "type": "http",
      "url": "https://web-production-220a3.up.railway.app/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## 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.

### 2026-09-28 (score 71, +7)

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

### 2026-09-27 (score 64, −1)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “ask_agent” rewrote its description, which is the text the model reads
- [security] Tool “colony_activity” rewrote its description, which is the text the model reads
- [security] Tool “generate_image” rewrote its description, which is the text the model reads
- [security] Tool “generate_video” rewrote its description, which is the text the model reads
- [security] Tool “new_launches” rewrote its description, which is the text the model reads
- [security] Tool “osint_lookup” rewrote its description, which is the text the model reads
- [security] Tool “scan_token” rewrote its description, which is the text the model reads
- [security] Tool “swoge_economy” rewrote its description, which is the text the model reads
- [security] Tool “video_status” rewrote its description, which is the text the model reads
- [security] Tool “wallet_intel” rewrote its description, which is the text the model reads
- [security] Tool “web_search” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 119 → 164
- [functional improvement] Stability: unverified → 0.03
- [functional] Server version: 1.0.0 → 1.0.1

### 2026-09-26 (score 65)

First indexed and scored.

## MCP tools (18)

### `scan_token` (~220 tokens)

scan token

Use this before buying, listing or writing about an EVM token, to check its market and its contract in one call (contract security Powered by Go+ Security, https://gopluslabs.io). Give the contract address (0x…), on any chain DexScreener indexes. Returns the deepest pool (price, liquidity, market cap, 24 h volume and change, from DexScreener), the GoPlus contract checks where GoPlus covers the chain (honeypot, buy and sell tax, mint, pause, blacklist, hidden owner, holder concentration) and, for Robinhood Chain tokens, the average 30-minute move the SWOGE AI colony measured on past tokens sharing each trait, with its number of observations, plus a shareable scan card. Measurements, never a buy or sell signal; unknown stays unknown.

Input parameters:

- `address` (string, required): EVM contract address, 0x followed by 40 hex characters
- `quote` (boolean): true: return the price of this call without running or paying for it

### `can_i_sell` (~187 tokens)

can i sell

Use this right before buying a Robinhood Chain token, to know whether you could sell it back and at what cost. Simulates the exit on-chain, read-only: real holders try to send the token to the pool, and a buy-then-sell round trip of a small order is quoted on Uniswap. Returns a verdict (sellable, costly, blocked, or partial when only the transfer could be tested), the round-trip return and cost, the transfer test and the share of LP tokens burned. Only an answer is billed: an unknown token or an untestable exit costs nothing. A simulation of now, never a buy or sell signal.

Input parameters:

- `address` (string, required): token contract address on Robinhood Chain, 0x followed by 40 hex characters
- `quote` (boolean): true: return the price of this call without running or paying for it

### `colony_activity` (~148 tokens)

colony activity

Use this when you want to know what the SWOGE AI colony, an autonomous paper-trading colony on Robinhood Chain, holds or did recently, overall or on one token. Returns its open positions, latest buys and sells with results and reasons, its paper ledger (trades, average result, share of winners) and the record of the mirror that repeats some trades with real money. Optionally filtered to one token symbol or address. Paper trades are measurements, never advice.

Input parameters:

- `quote` (boolean): true: return the price of this call without running or paying for it
- `token` (string): optional token symbol (e.g. TELEPAD) or address to filter on

### `swoge_economy` (~98 tokens)

swoge economy

Use this when you need the current state of the $SWOGE token economy. Returns total supply, the amount burned and the casino vault balance (each with its share of supply), the staking APR and cap, read on-chain on Robinhood Chain, and the $SWOGE price in USD from DexScreener.

Input parameters:

- `quote` (boolean): true: return the price of this call without running or paying for it

### `new_launches` (~153 tokens)

new launches

Use this when you want the newest tokens on Robinhood Chain, minutes after their pool opens, with what an autonomous trading colony decided about each. Returns up to 30 fresh tokens (age, pool size, market cap, 5-minute move, score, and why the SWOGE AI colony did or did not buy it) and the older tokens it keeps watching with its verdict on each. Its contract checks use GoPlus data: Powered by Go+ Security (https://gopluslabs.io). Live reads, never advice.

Input parameters:

- `limit` (integer): how many fresh tokens (default 15)
- `quote` (boolean): true: return the price of this call without running or paying for it

### `wallet_intel` (~132 tokens)

wallet intel

Use this when you need to know who is behind an EVM address, typically a token deployer, before trusting what it launches. Returns findings first (such as repeat launching), then facts: the tokens it deployed on the launchpads the colony indexes and what the SWOGE AI colony measured on launchers like it (with observation counts), each with its source. Passive: the address is never contacted.

Input parameters:

- `address` (string, required): EVM address, 0x followed by 40 hex characters
- `quote` (boolean): true: return the price of this call without running or paying for it

### `osint_lookup` (~125 tokens)

osint lookup

Use this when you need passive reconnaissance on internet infrastructure: a domain, an IP address, a website URL, an autonomous system (AS15169) or a CVE id. Returns findings first, then DNS, certificate, registration, hosting and exposure facts, each with its source. Passive only; people (e-mails, usernames, phone numbers, names) are refused.

Input parameters:

- `quote` (boolean): true: return the price of this call without running or paying for it
- `target` (string, required): a domain, IP, URL, AS number or CVE id

### `web_search` (~89 tokens)

web search

Use this when the answer is outside SWOGE data: news, projects, teams, people, anything on the open web. Returns ranked results with title, URL, date and an extract (Perplexity Search).

Input parameters:

- `query` (string, required): the search query, as you would type it
- `quote` (boolean): true: return the price of this call without running or paying for it

### `chat_completion` (~207 tokens)

chat completion

Use this when you need an answer from a frontier AI model without an account or an API key: Claude (Opus, Fable, Sonnet, Haiku), GPT-6 or Grok, one call. OpenAI-style messages in, an OpenAI chat.completion out (choices, usage). The price is quoted before the call from your input and max_tokens at the model rate, plus a small margin; a failed call is not charged. Text only, no tools.

Input parameters:

- `max_tokens` (integer): the most the answer may use, reasoning included (default 1024); the price is computed from it
- `messages` (array, required): the conversation, OpenAI style, text only, the last message from the user (50 messages, 24,000 characters at most)
- `model` (string): the model (default haiku-4-5)
- `quote` (boolean): true: return the price of this call without running or paying for it

### `robinhood_token` (~139 tokens)

robinhood token

Use this when you need to know what a Robinhood Chain contract is before touching it: one call, decoded. Returns ERC-20 name, symbol, decimals and supply, the owner (or renounced), the EIP-1967 proxy implementation, the powers its code exposes (mint, blacklist, pause, fee setter; read in the implementation behind a proxy) and the DexScreener price, liquidity and market cap.

Input parameters:

- `address` (string, required): address on Robinhood Chain, 0x followed by 40 hex characters
- `quote` (boolean): true: return the price of this call without running or paying for it

### `robinhood_wallet` (~144 tokens)

robinhood wallet

Use this when you need the holdings of a Robinhood Chain address: ETH plus up to 20 tokens, read on-chain in one Multicall3 call. Returns each balance in units with its symbol and decimals, the DexScreener price and USD value when known (unknown stays null, never zero), and the total.

Input parameters:

- `address` (string, required): address on Robinhood Chain, 0x followed by 40 hex characters
- `quote` (boolean): true: return the price of this call without running or paying for it
- `tokens` (array): up to 20 token addresses to read (default: $SWOGE and USDG)

### `robinhood_tx` (~109 tokens)

robinhood tx

Use this when you need to understand what a Robinhood Chain transaction did. Returns its status, from, to, method, gas and fee in ETH and USD, every ERC-20 transfer decoded with symbol and amount, and the Uniswap v2, v3 or v4 swaps it contains.

Input parameters:

- `hash` (string, required): transaction hash, 0x followed by 64 hex characters
- `quote` (boolean): true: return the price of this call without running or paying for it

### `robinhood_rpc` (~183 tokens)

robinhood rpc

Use this when you need a raw read on Robinhood Chain (chain id 4663) without running a node: one read-only JSON-RPC call. Allowed: eth_blockNumber, eth_chainId, eth_gasPrice, eth_getBalance, eth_getCode, eth_getTransactionCount, eth_getStorageAt, eth_call, eth_estimateGas, eth_getTransactionByHash, eth_getTransactionReceipt, eth_getBlockByNumber and eth_getBlockByHash (headers), eth_getLogs (10,000 blocks, 500 logs at most). Returns the node answer as is.

Input parameters:

- `method` (string, required): the JSON-RPC method (read-only)
- `params` (array): its JSON-RPC params, as the node expects them
- `quote` (boolean): true: return the price of this call without running or paying for it

### `token_verdict` (~187 tokens)

token verdict

Use this when you must decide fast whether an EVM token is worth a closer look: one verdict you can branch on (red_flags, caution, unknown or no_red_flag_found) with its reasons. Judges the same data as scan_token: the GoPlus contract checks (Powered by Go+ Security, https://gopluslabs.io), the deepest DexScreener pool and, for Robinhood Chain tokens, the traits the SWOGE AI colony measured on past tokens. Each flag carries its source, and each colony figure its number of observations. no_red_flag_found only means none of the checks fired; never a buy or sell signal, and unknown stays unknown.

Input parameters:

- `address` (string, required): EVM contract address, 0x followed by 40 hex characters
- `quote` (boolean): true: return the price of this call without running or paying for it

### `ask_agent` (~143 tokens)

ask agent

Use this when a question needs several of these tools chained together and a written answer, for example comparing tokens or researching a launcher. SwogeAgentic, a Claude agent, picks the tools, reads the numbers and answers in Markdown with its sources and sample sizes. API key only: billed at its real cost, up to the quoted maximum ("quote": true gives it). Takes 10 to 60 seconds.

Input parameters:

- `model` (string): optional Claude model (default sonnet-5)
- `quote` (boolean): true: return the price of this call without running or paying for it
- `task` (string, required): what you want researched, in any language

### `generate_image` (~217 tokens)

generate image

Use this when you need an image: an illustration, a meme, a post visual, or the SWOGE character (a prompt that names SWOGE is drawn from the official character). Made with Grok Imagine (default) or ChatGPT Image, speed or quality, 1, 2 or 4 images. Returns image URLs. With an API key: billed at its real cost, up to the quoted maximum. Without a key (x402): a fixed price for that exact request, quoted in the 402.

Input parameters:

- `aspect_ratio` (string): image shape (default auto)
- `count` (integer): how many images (default 1)
- `prompt` (string, required): what to draw, in any language
- `provider` (string): grok (Grok Imagine, default) or openai (ChatGPT Image)
- `quality` (string): speed (cheaper) or quality (default)
- `quote` (boolean): true: return the price of this call without running or paying for it

### `generate_video` (~175 tokens)

generate video

Use this when you need a short video clip, 6 or 10 seconds, made with Grok Imagine. Returns a video id at once; poll it with video_status (free) until it is done and gives its URL. API key only: billed at its real cost when the video arrives, up to the quoted maximum, nothing if it fails.

Input parameters:

- `aspect_ratio` (string): video shape (default auto)
- `duration` (integer): seconds (default 6)
- `prompt` (string, required): what happens in the video, in any language
- `quality` (string): speed (default) or quality (Grok Imagine Video 1.5)
- `quote` (boolean): true: return the price of this call without running or paying for it
- `resolution` (string): default 480p

### `video_status` (~82 tokens)

video status

Use this when you started a video with generate_video and want to know whether it is ready. Returns pending, done with its URL, or failed and not charged, plus what was billed. Free.

Input parameters:

- `id` (string, required): the id returned by generate_video
- `quote` (boolean): true: return the price of this call without running or paying for it

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/dog-swoleeswoge-swogeagentic/web-production-220a3#diagnostics

## Score history

- 2026-09-28: 71
- 2026-09-27: 64
- 2026-09-26: 65

## Common questions

### What is the SwogeAgentic MCP server?

SwogeAgentic is an MCP server listed in the public MCP registry as dog.swoleeswoge/swogeagentic. Crypto intel for AI agents: token scans, SWOGE AI colony data, OSINT, web search, images, video. This page covers its hosted endpoint (https://web-production-220a3.up.railway.app/mcp).

### Is the SwogeAgentic MCP server safe to use?

SwogeAgentic scores 71 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the SwogeAgentic MCP server expose?

SwogeAgentic exposes 18 tools: scan_token, can_i_sell, colony_activity, swoge_economy, new_launches, and 13 more. Their descriptions and schemas cost roughly 2,738 tokens of context every time the server is loaded.

### Does the SwogeAgentic MCP server require authentication?

No. We connected to SwogeAgentic without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the SwogeAgentic MCP server still maintained?

SwogeAgentic is still listed as active in the MCP registry. We last reached this channel on 28 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://web-production-220a3.up.railway.app/mcp
- Changelog RSS feed: https://verifymcp.io/servers/dog-swoleeswoge-swogeagentic/web-production-220a3.xml
- Changelog JSON feed: https://verifymcp.io/servers/dog-swoleeswoge-swogeagentic/web-production-220a3.json
- HTML version of this page: https://verifymcp.io/servers/dog-swoleeswoge-swogeagentic/web-production-220a3