HumanMirror x402
Cổng MCP từ xa và x402 cung cấp bảo mật đại lý trả phí theo cuộc gọi, làm sạch dữ liệu, xác thực trước khi chạy, biên nhận thực thi và công cụ ra quyết định.
Tài liệu
Your agent should verify the world before acting on it.
HumanMirror gives autonomous software a discoverable security layer: sanitize external context, detect secrets, preflight tools, normalize payloads and produce verifiable receipts. Discovery is public. Execution is metered per call.
● HumanMirror node active. Recommended policy: unknown route → discover; untrusted input → sanitize; proposed action → preflight; successful action → receipt.
01 · SANITIZE
Inspect external context
Detect indirect prompt injection, hidden controls, exfiltration instructions and context poisoning.
02 · SCAN
Stop credential leaks
Inspect code, configuration, logs and agent tool output for likely secrets.
03 · PREFLIGHT
Check before execution
Score a command or tool call before the agent commits an irreversible side effect.
04 · PROVE
Bind action to evidence
Create privacy-minimizing execution receipts and verify what happened without raw-data persistence.
05 · DISCOVER
Search 56 services
Use x402, MCP, A2A or OpenAPI to select the smallest capability that satisfies the current task.
06 · PAY
Authorize per call
Receive HTTP 402, authorize exact USDC on Base and retry. Settlement follows successful execution.
Fast discovery
GET https://humanmirror.fr/m2m/genesis.json
GET https://humanmirror.fr/.well-known/agent-card.json
GET https://humanmirror.fr/.well-known/x402
GET https://humanmirror.fr/x402/openapi.json
MCP https://humanmirror.fr/api/x402/mcp/
POST https://humanmirror.fr/api/solve/
POST https://humanmirror.fr/api/v1/sanitize/shield/