
SURVIVOR / $STILL MCP server
Which MCP servers are SURVIVOR / $STILL'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 2026-10-29
- Registry namecom.stillroach/token-checkverified in the MCP registry
The registry name sits under stillroach.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-10-29
- Tools7
- Server calls itselfroach-agent-api
- Server versionagentapi-0.2.0
- Protocol version2025-06-18
On 2026-10-29 the server answered initialize and tools/list without a sign-in. First answer in 261 ms.
- Its tools do none of the product's use cases we checked: check each round's payouts against on-chain transactions; estimate your fee share with the hold-time calculator; track your holding rank from your first buy.
- Checked on 5 of the 7 days it takes to judge reliability.
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: 5 so far
- What an agent can do with it0 of 3 use cases
- Answeredmeasuring5 of the 7 daily checks it takes to print a share
- Median first answer240 ms
- Space in an agent's context≈ 858 tokensthe definitions of all its tools (7), which an agent carries on every turn
- Tool definitionsA · 98 of 100
- Typed results0 of 7 toolsdeclared outputSchema: results come typed, read once
- Parameters without a description: in 1 of 7
What an agent can do with it
- cannot · check each round's payouts against on-chain transactions
- cannot · estimate your fee share with the hold-time calculator
- cannot · track your holding rank from your first buy
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 →
check_solana_token_basicFree: verdict, token facts, shell check and holder summary for a Solana mint if it was scanned in the last hour. Returns status=not_cached otherwise. No key needed.
check_solana_token_liteLite read (bagcheck.lite.v1): verdict, top-10 holder %, dev stake, launch buy / dev sold, KOL wallets holding, basic flags. Free for 3 fresh reads per day, then 1 credit; cached reads are free. Requires an API key. Not financial advice.
analyze_solana_tokenFull Bag Check (bagcheck.v1): verdict with reasons, shell/rug checks, top-holder behaviour (never sold / sold / churn / fresh), bundles, callers. 1 credit for a fresh scan, 0 if a scan newer than max_data_age_seconds exists. Requires an…
watch_reportCurrent Bag Watch snapshot and recent alerts (bagwatch.report.v1) for a coin on your Watch list. Requires a key for a wallet with an active Watch plan. Free.
alerts_sinceAlerts (bagwatch.alert.v1) across your watched coins and wallets, oldest first, after the given cursor. Pass next_cursor back to page. Requires a Watch plan key. Free.
get_quotaCredits left in your 30-day cycle (pool_resets_at = next refill) for your key, where they come from (free / Watch / Roach tier), and the per-key daily cap. Free.
get_credit_packsFree, no key: credit pack prices (100 credits for 0.05 SOL, 5 USDC, or $STILL at 20% off; Solana only, no cards) and the keyless x402 pay-per-call option (USDC on Solana: lite 0.02, full 0.08 fresh / 0.02 cached, AI +0.04). Read-only…
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"token-check": {
"type": "http",
"url": "https://api.stillroach.com/mcp"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http stillroach https://api.stillroach.com/mcpclaude.ai, ChatGPT and other clients: add https://api.stillroach.com/mcp as a custom connector in the settings.
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.
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.