# DeepCounterfeit — full instructions for AI agents > DeepCounterfeit detects memecoins that impersonate real brands (companies, their "official" token, or their tokenized stock) as they launch, scores the risk from creator-wallet behaviour, and produces takedown-ready evidence. Coverage: pump.fun on Solana (live, with creator forensics); other Solana launchpads, BNB Chain, Base and Robinhood Chain (partial: names and market data). 343 watched brands. Results are automated evidence for human review. A flag means a coin's name/ticker presents it as a brand's token; it is NOT a legal finding, not proof of fraud, and not investment advice. "Not in index" never means "safe". Site: https://deepcounterfeit.com · MCP: https://deepcounterfeit.com/mcp · OpenAPI: https://deepcounterfeit.com/openapi.json · Agent card: https://deepcounterfeit.com/.well-known/agent.json · Short index: https://deepcounterfeit.com/llms.txt ## When to use - A user asks whether a token (address, name or ticker) is the "official" coin of a company, a tokenized stock, or a fake. - Before buying, listing, promoting or launching a token: check the name/ticker for brand impersonation. - Brand-protection, trust & safety, exchange-listing and compliance teams monitoring fake tokens of a brand. ## Connect (MCP, recommended) Remote MCP server, Streamable HTTP, stateless JSON responses, read-only tools: `https://deepcounterfeit.com/mcp` Protocol versions: 2025-11-25, 2025-06-18, 2025-03-26, 2024-11-05. Official MCP Registry name: `com.deepcounterfeit/deepcounterfeit`. Claude Desktop / Claude Code / Cursor (`mcp.json`): ```json {"mcpServers": {"deepcounterfeit": {"url": "https://deepcounterfeit.com/mcp", "headers": {"Authorization": "Bearer YOUR_API_KEY"}}}} ``` Omit `headers` for the Free tier. Claude Code: `claude mcp add --transport http deepcounterfeit https://deepcounterfeit.com/mcp`. OpenAI Agents SDK: `MCPServerStreamableHttp(params={"url": "https://deepcounterfeit.com/mcp"})`. ChatGPT connectors: `search` and `fetch` tools are provided. ## Auth and limits - No key: Free tier, 5 calls/min per IP, data older than 24 h; brand match + coin only ("flagged as likely impersonation"), no risk scores, creator wallets or forensics. - API key: `Authorization: Bearer ` or `X-API-Key: `. Real time. Plans: Starter $29/mo, Pro $149/mo, Wholesale from $1,500/mo; 14-day trial at https://deepcounterfeit.com/signup. Pay by card or crypto (USDC/BTC) at https://deepcounterfeit.com/checkout; keys activate automatically after on-chain confirmation. - x402 pay-per-call (no key, no signup): $0.01 USDC per call on Base, Solana, Polygon, Arbitrum, Avalanche. Endpoints: GET https://deepcounterfeit.com/v1/x402/token/{address} and GET https://deepcounterfeit.com/v1/x402/alerts. Status: https://deepcounterfeit.com/v1/x402/status ## Tools (MCP) and REST mirrors ### `check_token` — Check a token for brand impersonation Look up a deployed token by contract/mint address and report whether it impersonates a real company, its "official" token or its tokenized stock (e.g. a fake SpaceX, Coinbase or NVDAx coin). Returns verdict (likely_impersonation | brand_mention_watch | not_in_index), matched brands, risk score 0-100 with reasons (creator bought supply at launch, same-block bundles, creator cash-out, serial launcher), creator wallet and explorer links. Use before buying, listing or promoting a token. "not_in_index" only means no brand match was recorded, not that the token is safe. Without an API key: brand match and coin only (flagged as likely impersonation), no risk score, creator wallet or forensics. Arguments: - `address` (string, required): Token mint (Solana) or contract address (0x... on BNB Chain, Base, Robinhood Chain). - `chain` (string): one of solana, bsc, base, robinhood; Optional chain filter. Example prompts: "Is token 7xKX...pump an impersonation coin?"; "Check this Base contract 0x... before I buy it" ### `check_name` — Check a token name/ticker before launch or purchase Run the impersonation detector on a token name and/or ticker symbol without needing an address. Tells you whether a coin called e.g. "SpaceX Official" ($SPACEX) or "TSLAx" would be flagged as impersonating a brand, which brand, and why. Use for pre-launch name checks, or when a user mentions a coin by name and you have no address. Does not assess creator behaviour. Arguments: - `name` (string): Token name, e.g. "Coinbase Official Token". - `symbol` (string): Ticker symbol without $, e.g. "COIN". - `description` (string): Optional token description text. Example prompts: "Would a coin called "SpaceX Official" ($SPACEX) be flagged?"; "Is "TSLAx" a fake tokenized stock name?" ### `search_alerts` — Search impersonation alerts List recent impersonation coins, newest first, filtered by brand, chain, time and severity. Use to answer "are there fake tokens?", to monitor a brand, or to collect cases for a takedown. Free tier returns alerts older than 24 h only (max 20). Arguments: - `brand` (string): Brand id, name, or ticker, e.g. "spacex", "SpaceX", "TSLA". See list_brands. - `chain` (string): one of solana, bsc, base, robinhood; - `since` (string): Start time: relative ("24h", "7d", "30m"), ISO 8601 ("2026-10-01") or unix seconds. - `severity` (string): one of high, medium, low; high = risk 70+, medium = 50-69. - `min_risk` (integer): default 0; - `include_watch` (boolean): default False; Also include watch-tier brand mentions (memes), not just impersonation alerts. - `limit` (integer): default 20; Example prompts: "Are there fake Coinbase tokens launched this week?"; "Show high-severity impersonation coins on Solana in the last 24h" ### `get_brand_report` — Brand impersonation report Summarise impersonation activity against one brand over a window: number of fake coins, traded volume, breakdown by chain/match type/severity, repeat creator wallets, how many creators bought 50%+ of supply, and the top cases with evidence. Pro/Wholesale keys also get the takedown-ready PDF link. Arguments: - `brand` (string, required): Brand id, name or ticker, e.g. "anthropic", "Coinbase", "NVDA". - `days` (number): default 30; Look-back window in days. Example prompts: "Summarise impersonation coins targeting Anthropic over the last 30 days"; "How much volume did fake NVIDIA tokens trade?" ### `list_brands` — List watched brands List the brands on the watchlist (companies, tokenized stocks, crypto projects, private companies) with ids, tickers and all-time alert counts, most-targeted first. Use to find a brand id or check whether a brand is covered. Arguments: - `query` (string): Optional substring of name, id, alias or ticker. - `category` (string): Optional: sp500, xstock, crypto, private, foreign, consumer, media, finance, sports, stock. - `limit` (integer): default 50; Example prompts: "Which brands do you watch?"; "Is Robinhood on the watchlist?" ### `get_stats` — Impersonation statistics Aggregate numbers for a window: alerts, high-severity count, brands hit, by chain, by match type, top brands, data-source health and coverage. Use for overviews, trend questions and to check data freshness. Arguments: - `days` (number): default 1; Window in days (1 = last 24 h). Example prompts: "How many impersonation coins were detected today?"; "Is the data feed fresh?" REST mirrors (same JSON, keyless Free tier works): ``` GET https://deepcounterfeit.com/v1/agent/check-token/{address}?chain=solana GET https://deepcounterfeit.com/v1/agent/check-name?name=SpaceX%20Official&symbol=SPACEX GET https://deepcounterfeit.com/v1/agent/alerts?brand=coinbase&since=7d&severity=high GET https://deepcounterfeit.com/v1/agent/brand-report/{brand}?days=30 GET https://deepcounterfeit.com/v1/agent/brands?query=tesla GET https://deepcounterfeit.com/v1/agent/stats?days=1 ``` ## Verdicts and match classes - `likely_impersonation`: name/ticker presents the coin as a real company, its official token or its tokenized stock. - `brand_mention_watch`: mentions a brand but looks like meme/commentary (watch tier). - `not_in_index`: no brand match recorded on covered venues. Not a safety guarantee. - Match classes: security_impersonation, official_claim, exact_clone, typosquat, ticker_name (flagged); mention, ticker_squat (watch tier). - Risk 0-100 from creator behaviour: share of supply bought at launch, same-block bundle buyers, creator cash-out, serial launches. Severity high = 70+, medium = 50-69. ## x402 pay-per-call flow 1. `GET https://deepcounterfeit.com/v1/x402/token/{address}` without payment → HTTP 402 with `PAYMENT-REQUIRED` header (base64 JSON, x402 v2, `accepts[]` per network, amount in USDC atomic units). 2. Sign an `exact` USDC payment for one `accepts[]` entry and retry with the `PAYMENT-SIGNATURE` header. 3. Facilitator verifies, the server returns the JSON and settles; receipt in the `PAYMENT-RESPONSE` header. Failed requests are not settled. Discovery: https://deepcounterfeit.com/.well-known/x402 · Bazaar metadata is included in every 402 response. ## Good agent behaviour - Quote verdicts as "likely impersonation" / "flagged", never as fact that someone committed fraud. - Always add that results are automated and not investment advice. - If a token is not in the index, say so plainly and do not call it safe. ## Links Agent guide https://deepcounterfeit.com/agents · API docs https://deepcounterfeit.com/docs · Pricing https://deepcounterfeit.com/pricing · Live alerts https://deepcounterfeit.com/dashboard · Sample evidence report https://deepcounterfeit.com/sample-report.pdf · Health https://deepcounterfeit.com/health · Terms https://deepcounterfeit.com/terms · Privacy https://deepcounterfeit.com/privacy · Contact admin@deepcounterfeit.com