crosscheck

Endpoints (10 live)

• POST /v1/skillcheck — Security scan of an AI agent skill or MCP server before install: a malware and prompt injection audit of its files (read, never run). Finds downloads piped to a shell, credential and wallet theft, secrets sent over the network, persistence, hidden Unicode, and instructions aimed at the agent or the scanner. Supply-chain check with a signed receipt keyed to the file hashes; bundles already scanned are free to look up. $0.03. (0.03 USDC on Base)
• POST /v1/accept — Verify output and validate work another AI agent hands back, before you pay, release escrow, or pass it on. An acceptance test and quality check: send the task and the deliverable, get accept or reject with every requirement judged. Counts, required JSON fields, and sums are checked in code. Deliverables that certify themselves or inject instructions never pass. Signed receipt binding task, deliverable, and payment hashes. $0.03. (0.03 USDC on Base)
• POST /v1/check — Fact-check and review an AI agent's draft before a human sees it: a hallucination check and second opinion on any email, report, summary, or PR description. Returns pass or specific issues with fixes. Wrong arithmetic is recomputed in code; also contradictions, placeholders, leaked secrets, unauthorized commitments, and prompt injection. Send sources to verify citations and claims against them. Signed receipt. $0.02. (0.02 USDC on Base)
• POST /v1/accept-lite — Verify output another AI agent hands back, by code alone, for $0.005: no model call, answers in about a second. Recomputes the deliverable's written sums, totals, percentages, dates, weekdays, time spans, unit conversions, and list counts; checks lengths the task asks for; flags placeholders and prompt injection; measures words, items, and JSON shape. Signed receipt, tier lite. For every requirement judged, use accept ($0.03). (0.005 USDC on Base)
• POST /v1/skillcheck-lite — Security scan of an AI agent skill or MCP server by code rules alone, for $0.005: no model call. Reads every file (never runs them) for downloads piped to a shell, credential and wallet theft, secrets sent out, persistence, hidden Unicode, text aimed at the scanner, and packages missing from or removed by their registry. Signed receipt, tier lite. For a model reading too, use skillcheck ($0.03). (0.005 USDC on Base)
• GET /v1/skillcheck-lite (free)
• GET /v1/accept-lite (free)
• GET /v1/skillcheck (free)
• GET /v1/check (free)
• GET /v1/accept (free)

MCP tools (6)

crosscheck https://crosscheckapi.com/mcp

• accept — Checks work another agent handed back against the task it was given, before you pay, release escrow, or pass it on. Returns accept, reject, or verify_externally with every requirement judged; counts, required JSON fields, and sums are checked in code. Optional payment_tx is verified on chain. $0.03 base. Paid per call in USDC over x402 (Base, or free test USDC on Base Sepolia). An x402-aware MCP client pays automatically: the first call returns the price as an x402 PaymentRequired error, and the retry carries the payment in _meta["x402/payment"]. Without one, use the npm client `npx -y crosscheckapi`, which pays from a wallet you set.
• check — Independent review of a draft (email, report, summary, PR description) before its human sees it. Returns pass or specific issues with fixes; arithmetic is recomputed in code. With request (what the human asked), it also checks the draft answers that. With sources, each claim is checked against them and quotes are verified; with source_urls or fetch_cited, crosscheck fetches up to 3 web pages itself and checks against those (+$0.01 a page, and a dead link the draft gives fails it). $0.02. Paid per call in USDC over x402 (Base, or free test USDC on Base Sepolia). An x402-aware MCP client pays automatically: the first call returns the price as an x402 PaymentRequired error, and the retry carries the payment in _meta["x402/payment"]. Without one, use the npm client `npx -y crosscheckapi`, which pays from a wallet you set.
• result — Free. Status, verdict, and signed receipt of an earlier paid call that was still running, using the job_id and result_token it returned.
• skillcheck — Security review of a skill or MCP server's files before you install or connect it: read, never run. Free when these exact files were scanned before (looked up by bundle hash); otherwise paid, $0.03 base. Never says safe. Paid per call in USDC over x402 (Base, or free test USDC on Base Sepolia). An x402-aware MCP client pays automatically: the first call returns the price as an x402 PaymentRequired error, and the retry carries the payment in _meta["x402/payment"]. Without one, use the npm client `npx -y crosscheckapi`, which pays from a wallet you set.
• skillcheck_lookup — Free. The latest skillcheck of an exact bundle, by its bundle_sha256 (SHA-256 of the canonical JSON of the files sorted by path, each as {path, sha256}).
• verify_receipt — Free. Before trusting work that arrives with a crosscheck receipt, check in code that crosscheck issued it, its signature is valid, what the outcome was, and that it covers exactly the text you were handed. Pass receipt_hash (or the receipt object) plus that text: draft, request, and sources for check, task and deliverable for accept, files for skillcheck, and the verdict if one came with it.