Skip to main content
Glama

YieldAgentX402 SDKs

Public client libraries and integrations for YieldAgentX402 — the custody-free, policy-gated, receipt-backed execution layer for AI agents.

Platform · MCP Docs · Try without signup · Get API key

yieldagentx402-sdks MCP server

This repo hosts the open-source distribution packages. The platform itself (gateway, ShadeGuard policy engine, TEE attestation, orchestrator, Filecoin/BTFS anchoring) lives in a separate private repository — we ship the bits agents need to use the platform here.


Packages in this repo

Package

Registry

What it does

sdk-ts/

npm: yieldagentx402

Typed TypeScript SDK. Wraps all 18 MCP tools with first-class types, idempotency, webhook verify, Shade Agent helpers. Zero deps. Works in Node, browsers, Workers, Deno, Bun.

mcp-server/

npm: agentx402-mcp-server

Stdio MCP wrapper. Drop into Claude Desktop / Code / Cursor / Windsurf for all 18 YieldAgentX402 tools across 18 chains. Glama: listing.

verify-lib/

npm: yieldagentx402-verify

Zero-dependency WebCrypto verifier for receipts + webhook signatures. Works in browser, Node 18+, Cloudflare Workers, Deno.

crewai-tools/

PyPI: crewai-yieldagentx402-tools

CrewAI tool wrappers — policy dry-run, x402 payment, secure workflow, receipt verify, Shade Agent MPC wallet status.

langchain-tools/

PyPI: yieldagentx402-langchain

LangChain tool wrappers — discovery, policy preview, receipt verify, action planner (MCP gateway).


Related MCP server: dingdawg-governance

Try without signup

curl -X POST https://api.yieldagentx402.app/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call",
       "params":{"name":"yax_get_capabilities","arguments":{}}}'

No API key required. Returns all 18 tools, 18 chains, rate limits, fees, supported runtime configs.

Want an instant test key?

curl -X POST https://api.yieldagentx402.app/api/apply \
  -H "Content-Type: application/json" \
  -d '{"agent_name":"my-agent"}'

Returns a yax_test_* key (5 USD cap, 7-day expiry) for safe read-only tools.


What is YieldAgentX402?

A trust layer for agents that move money.

  • 18 MCP tools — discovery, secure workflow execution, x402 payment settlement, policy dry-run, approvals, receipt verification, attestation, finance/CRM skill tools

  • 18 chains — Base, Ethereum, Bitcoin (native), Starknet, NEAR, Solana, Stacks, BNB, Rootstock, Filecoin EVM + native, Aptos, Sui, TON, Tron, XRPL, Stellar, Algorand

  • Custody-free wallet — one NEAR MPC key authority derives EVM + BTC addresses on demand. No private keys held by the gateway or worker. Starknet uses Stark curve preset.

  • Intel TDX TEE attestation — mrEnclave verifiable at /api/tee/report

  • HMAC-SHA256 receipts — every call returns a signed receipt

  • Filecoin + BTFS dual anchoring — permanent tamper-proof proof storage

  • x402 protocol settlement — 0.25% protocol fee, machine-verifiable in /.well-known/x402 as extra.protocolFeeBps: 25


Custody-free wallet — one MPC key, three chains

Chain

Address

Curve

EVM (Base/Eth)

0x6905D04C3655625F6171f7b1e801a9325B1Fa1e8

secp256k1 (MPC)

Bitcoin (P2WPKH)

bc1qcd8ljsj7x0rq3sw7drrt7ragqmmt3tfsfxs6gy

secp256k1 (MPC)

Starknet

0x003a4e08…570906a21

Stark (preset)

Live manifest: https://shade-agent-worker.cryptoblac.workers.dev/api/wallet


Live infra

Surface

URL

MCP endpoint

https://api.yieldagentx402.app/mcp

Public discovery

POST /mcp with yax_get_capabilities (no auth)

Instant test key

POST /api/apply

Ready-to-paste configs

GET /api/agent-onboard

x402 manifest

https://api.yieldagentx402.app/.well-known/x402

MCP manifest

https://api.yieldagentx402.app/.well-known/mcp.json

Status

https://api.yieldagentx402.app/api/status

TEE report

https://api.yieldagentx402.app/api/tee/report

Shade Agent wallet

https://shade-agent-worker.cryptoblac.workers.dev/api/wallet

Receipt verifier (UI)

https://yieldagentx402.app/verify

Postman collection

https://yieldagentx402.app/postman/yieldagentx402.postman_collection.json


Contributing

Issues and PRs on the public SDK packages are welcome here. Platform-level changes (gateway, policy engine, TEE worker) happen in a private repo — open an issue if you have a feature request and we'll triage.


License

MIT — see LICENSE.

Available Tools

18 tools
yax_audit_complianceA
Read-onlyIdempotent

Audits recent agent activity against your policy rules and flags violations. Returns compliant vs flagged run counts, violation details, and a Filecoin-anchored compliance report CID.

Inputs: period, policy_id, format. Returns: ok, run_id, compliant_runs, flagged_runs, +2 more

yax_build_email_sequenceA

Builds a personalized cold email sequence for a list of prospects. Researches each prospect and writes tailored copy. Returns subject lines, body, and send schedule per prospect.

Inputs: prospects (required), offer (required), sequence_length, tone. Returns: ok, run_id, sequences

yax_check_policyA
Read-onlyIdempotent

Checks if an action would be approved by your policy without running it. Returns approved/denied, remaining spend cap, and whether human approval is required.

Inputs: action_type (required), parameters. Returns: decision, policy_version, shade_guard, cap_remaining_usd, +1 more

yax_collect_ar_invoicesA

Analyzes open invoices by age (30/60/90+ days) and sends escalating follow-up emails to overdue clients. Returns invoice count, total overdue amount, and emails sent per bucket.

Inputs: aging_buckets, send_emails, escalate_at. Returns: ok, run_id, invoices_scanned, overdue_count, +3 more

yax_control_workflowB
Destructive

Stop (cancel) or resume (approve human gate) an orchestration workflow run for this agent.

Inputs: workflow_id (required), action (required).

yax_enrich_lead_dataA

Parses a raw lead email, form submission, or webhook payload into a structured contact record, enriches it, and logs it to your CRM. Returns name, email, company, phone, and ICP score.

Inputs: source (required), raw_content, crm_push. Returns: ok, run_id, lead, crm_synced

yax_forecast_cash_flowA
Read-onlyIdempotent

Generates a week-by-week cash flow forecast across base, optimistic, and pessimistic scenarios. Returns inflow/outflow projections and a Filecoin-anchored report CID for board or lender use.

Inputs: horizon_weeks, scenarios, include_ar. Returns: ok, run_id, forecast, report_cid

yax_get_attestationA
Read-onlyIdempotent

Checks whether the TEE enclave is operational. Returns enclave status, the verifiable mrEnclave hash, and timestamp of last attestation. Use before high-value actions.

Inputs: no named parameters. Returns: ok, status, enclave_hash, last_attestation, +1 more

yax_get_capabilitiesA
Read-onlyIdempotent

Returns every tool available on this server, what each one does, and when to use it. Call this first when connecting to understand the platform.

Inputs: no named parameters. Returns: differentiators, security, tool_categories, getting_started

yax_get_receiptA
Read-onlyIdempotent

Fetches the full signed receipt for any past run. Returns policy result, TEE attestation, input/output hashes, and Filecoin + BTFS proof CIDs.

Inputs: run_id (required). Returns: run_id, policy, tee_attestation, input_hash, +4 more

yax_get_wallet_statusA
Read-onlyIdempotent

Returns the agent's wallet balance, spend cap, remaining cap for this period, and wallet address. Use before a payment to confirm funds are available.

Inputs: no named parameters. Returns: balance_usdc, spend_cap_usd, cap_remaining_usd, custody

yax_list_runsA
Read-onlyIdempotent

Lists recent agent runs with status and policy result. Filter by outcome (approved/denied/escalated) to audit activity or find run IDs for receipt lookup.

Inputs: limit, outcome, since. Returns: runs, total

yax_monitor_churn_riskA

Flags customers at risk of churning based on usage and payment patterns, and drafts a re-engagement email per account. Returns a prioritized risk list with churn score and outreach copy.

Inputs: customer_ids, risk_threshold, draft_emails. Returns: ok, run_id, at_risk, total_flagged

yax_process_x402_paymentA
Destructive

Sends an x402 payment to any EVM, NEAR, or Stacks address. Validates against your spend cap before signing. Returns transaction hash and signed receipt.

Inputs: recipient (required), amount (required), token, memo. Returns: run_id, tx_hash, status, proof, +2 more

yax_run_secure_workflowA
Destructive

Runs any agent action — email, API call, on-chain tx, or workflow — with policy enforcement and TEE attestation. Returns a run ID, policy decision, and signed Filecoin-anchored receipt.

Inputs: intent (required), action_type (required), parameters, spend_cap_override, +1 more. Returns: run_id, status, policy, tee_attestation, +3 more

yax_score_leadsA

Scores inbound leads by conversion likelihood and syncs enriched records to your CRM. Returns each lead with a score (0–100), tier (hot/warm/cold), and enrichment data.

Inputs: leads (required), crm_push, score_model. Returns: ok, run_id, leads, crm_synced

yax_submit_approvalB

Approve or deny a pending policy approval token for the authenticated agent.

Inputs: approval_id (required), decision (required).

yax_verify_receiptA
Read-onlyIdempotent

Confirms a receipt's cryptographic signature is valid and its Filecoin anchor exists. Returns verified: true/false, signature status, and anchor CIDs.

Inputs: run_id (required), filecoin_cid. Returns: verified, signature_ok, anchor_exists, filecoin_cid, +1 more

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 18 tools

    Disambiguation4/5

    Most tools have distinct purposes (e.g., audit vs. email sequence vs. payment), but slight overlap exists between yax_enrich_lead_data and yax_score_leads (both deal with leads) and between yax_run_secure_workflow and yax_control_workflow (both manage workflows). Descriptions clarify differences.

    Naming Consistency5/5

    All tools follow the consistent pattern 'yax_verb_noun' in snake_case, e.g., yax_check_policy, yax_forecast_cash_flow. No mixing of styles or irregular names.

    Tool Count5/5

    18 tools is appropriate for a multifunctional agentic hub covering audit, email, invoices, leads, workflow, payments, and attestation. Each tool serves a clear niche without redundancy.

    Completeness4/5

    Covers major business operations (CRM, invoicing, payments, compliance, workflow), but lacks tools for direct policy editing or advanced CRM management. Minor gaps exist for a fully autonomous agent.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      D
      maintenance
      Universal governance layer for AI agents — MCP-native, fail-closed, LNN interpretability. Governed receipts, IPFS audit proofs, and rollback for any agent in any framework.
      3
      78 npm
      Apache 2.0
    • A
      license
      B
      quality
      B
      maintenance
      A governance proxy for AI tools — every MCP/agent tool call is policy-gated, secret-redacted, and written to a hash-chained, offline-verifiable audit trail.
      13
      MIT