{
  "markdown": "# @mindjack/mcp\n\n[![npm](https://img.shields.io/npm/v/%40mindjack%2Fmcp)](https://www.npmjs.com/package/@mindjack/mcp)\n\nSolana memecoin rug check and token risk for trading agents, as MCP tools: pump.fun launches and migrations, calibrated rug probability, sniper / insider / bundle and fresh-wallet detection, holder clusters, KOL trades, wallet history across every launch indexed, and a live sellability check before you buy. Built for the trenches.\n\n## Install\n\nAdd to your MCP client config — Claude Desktop, Cursor, Codex, or anything else\nthat speaks MCP:\n\n```json\n{\n  \"mcpServers\": {\n    \"mindjack\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@mindjack/mcp\"]\n    }\n  }\n}\n```\n\nThat is the whole setup. No signup, no card: the server mints a key on first\nuse, saves it to `~/.config/mindjack/key`, and reuses it from then on. A new\nkey starts empty — fund it with a USDC deposit when you are ready; the free\ntools below work before you fund anything, and `get_sample` answers one real\ntoken in full so you can see every paid tier before spending a cent.\n\nSet `MINDJACK_API_KEY` to use a specific key instead, or `MINDJACK_CONFIG_DIR`\nto keep it somewhere else.\n\nPrefer to skip keys entirely? The API also speaks\n[x402](https://x402.org) — pay per call in USDC on Solana, no account at all.\n\n## Tools\n\n| Tool | Price | Answers |\n|---|---|---|\n| `get_sample` | free | One fixed token, answered in full — no key, no payment |\n| `get_coverage` | free | What we hold, how fresh, and a mint that works |\n| `get_scorecard` | free | Our measured hit rate per risk band |\n| `get_balance` | free | What is left on your key |\n| `check_token` | $0.001 | Is this one dangerous, and how fast do these collapse |\n| `find_tokens` | $0.005 | Which tokens are worth looking at, verdicts attached |\n| `search_tokens` | $0.005 /page | Find any token we ever analysed, by symbol, name or mint |\n| `search` | $0.005 /page | The same search, shaped for research clients: `{id, title, url}` rows to cite or pass to `fetch` |\n| `inspect_token` | $0.005 | Who is holding it, and what kind of wallets |\n| `token_wallets` | $0.005 | The named wallets behind it: insiders, snipers, fresh, wash, KOLs — with funders and clusters |\n| `token_price_path` | $0.005 | What it did after we called it |\n| `can_i_exit` | $0.005 | Can it be sold right now — a live Jupiter round trip at $100 and $1000 |\n| `check_wallet` | $0.006 | Who is this wallet, across every token we indexed |\n| `kol_leaderboard` | $0.02 /page | Every tracked KOL ranked: realized SOL, win rate, recent window |\n| `kol_record` | $0.02 | One KOL's full record: per-token history and latest trades |\n| `token_identity` | $0.025 | Who is behind it, and how their earlier tokens ended |\n| `token_report` | $0.025 | Everything on one token in one call; `depth: \"full\"` is $0.07 |\n| `fetch` | $0.025 | The same report as one document, keyed by the `id` a `search` result gave you |\n| `token_changes` | $0.025 | Who sold since we analysed it (reads the chain now) |\n| `test_hypothesis` | $0.025 | What happened to tokens shaped like this |\n| `find_serial_insiders` | $0.025 / 25 | Who keeps turning up as an insider, across the whole index |\n| `compare_tokens` | $0.025 | Were these 2-4 tokens run by the same people |\n| `wallet_network` | $0.025 | Who a wallet moves with: counterparts, direction, second hop |\n| `funder_networks` | $0.025 /page | Funders seeding fresh wallets index-wide, exchanges named |\n| `token_graph` | $0.04 | How the holders are connected: edges, clusters, wash wallets |\n| `token_web` | $0.04 | Launches tied to this one through shared wallets, with outcomes and peaks |\n\n`search` and `fetch` are aliases, not extra products: same endpoints, same\nprices, same billing as `search_tokens` and `token_report`. They exist under\nthose exact names because ChatGPT's deep research and company knowledge modes\nlook for them by name and connect to nothing without them.\n\nLive feeds (new launches, watched wallets, KOL trades) are server-sent event\nstreams and do not map to MCP tools — the endpoint guide below covers them.\n\n## Resources\n\nFour read-only documents, all free, attached once rather than called per task.\nA client can cache them; the model does not have to decide to fetch them.\n\n| URI | What it is |\n| --- | --- |\n| `mindjack://coverage` | The window we hold, what is in it, and a live mint guaranteed to have data |\n| `mindjack://scorecard` | Every calibrated band with the collapse rate measured for it and its sample size |\n| `mindjack://prices` | Every priced route with its price, asset, network and receiving address |\n| `mindjack://sample` | One real token answered in full, free, so you can read the shape before buying it |\n\n## Prompts\n\nThree workflows, written down. Each is the order of calls that answers a real\nquestion, with what to read out of each step and when the next one is worth its\nprice.\n\n| Prompt | Arguments | Answers |\n| --- | --- | --- |\n| `vet_before_buying` | `mint` | Should I take a position in this token? |\n| `find_candidates` | `hours`, `min_mcap` (optional) | What launched recently that is worth a closer look? |\n| `vet_counterparty` | `address` | Who is this wallet and what has it done before? |\n\nMCP passes prompt arguments as strings, so send `\"6\"` rather than `6`.\n\n## Using it\n\nEvery tool declares an `outputSchema` and returns `structuredContent` beside\nthe text, so a client can validate the answer and plan the next call from the\nfield names without paying to discover the shape.\n\nA first run, in the order the server itself recommends:\n\n```\nget_coverage                      -> free; the window, and a mint that works\nfind_tokens  hours=24 max_rug_pct=45   -> candidates, each with a measured verdict\ncheck_token  mint=<from above>    -> $0.001; structure and collapse speed\ntoken_identity mint=<survivor>    -> $0.025; who is holding it and how their\n                                     other tokens ended\n```\n\n`max_rug_pct` is worth one warning: it is a **measured** collapse frequency for\na calibrated band, not a score that starts at zero. The safest band we publish\nstill rugged about 35% of the time, so a threshold under that matches nothing\nat any window length. The universe base rate is 45%.\n\n## What makes this different\n\n**We keep outcomes.** Most APIs describe a token as it looks right now. We also\nknow what happened to every launch we have indexed since March 2026 — tens of\nthousands of them — which is why `check_token` can tell you that tokens scoring\nlike this one collapsed 78% of the time rather than just handing you a number.\n\n**We publish our hit rate.** `get_scorecard` returns the measured accuracy of\nevery band, refreshed weekly. Nobody who does not keep labelled outcomes can\nproduce that, and you should not trust a risk score from anyone who won't.\n\n**Collapse speed, not just collapse probability.** In this market almost\neverything eventually dies, so \"will it\" separates weakly. \"How fast\" separates\nabout tenfold — the worst band's median collapse is 19 seconds from peak, the\ncleanest is 11 minutes. That is the number that decides whether a position is\nexitable at all.\n\n**Wallets have a past.** Millions of recorded wallet appearances. When the same wallets\nshow up in a new launch, `token_identity` tells you where else they have been\nand how those ended.\n\n## Honest limits\n\n- **Solana only**, and only tokens indexed at migration. We hold every pump.fun\n  and letsbonk migration since our start date — complete inside that window,\n  nothing before it. The record is point-in-time and cannot be rebuilt later.\n- **We do not predict price.** Where upside is reported it is a measured\n  historical frequency for a cohort, not a forecast, and not adjusted for fees\n  or slippage.\n- **Unknown tokens, failed calls and empty results cost nothing.** Every\n  response carries `_meta.coverage` and `_meta.billing` so you can see exactly\n  what you got and what it cost.\n\n## Links\n\n- API: <https://api.mindjack.xyz>\n- Endpoint guide an LLM can read in one fetch: <https://api.mindjack.xyz/llms.txt>\n- OpenAPI: <https://api.mindjack.xyz/v1/openapi.json>\n- Source, issues and releases: <https://github.com/freyotrisolana/mindjack-mcp>\n- Listings: [npm](https://www.npmjs.com/package/@mindjack/mcp), [MCP Registry](https://registry.modelcontextprotocol.io/v0/servers/xyz.mindjack%2Fmcp/versions), [Smithery](https://smithery.ai/servers/mindjack/intel), [Glama](https://glama.ai/mcp/servers/freyotrisolana/mindjack-mcp), [Cursor Directory](https://cursor.directory/plugins/mindjack), [verifymcp](https://verifymcp.io/servers/xyz-mindjack-mcp/api)\n",
  "bytes": 8592,
  "sha": "311003e9b0b1a94ed5e79fb5194bff30fbc5045e67e672f53cbc8db6f2d405f9",
  "repo_slug": "freyotrisolana/mindjack-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_xyz_mindjack_mcp_24c1d2b0/readme"
}