MCP service candidate
Social Intel API
Inspect its declared service, tool surface, and current score before connecting it to an agent client.
Inspect agentCodex
MCP lets Codex reach external tools, data, and workflows through the Model Context Protocol. Use it when a coding task needs a repeatable capability outside the repo: documentation lookup, issue access, database inspection, API calls, or another agent. Use The Spawn when that capability is an ERC-8004 onchain agent with inspectable identity, endpoint, and payment signals.
Named task: docs lookup, issue triage, analytics query, API call, or paid data fetch
Official or trusted publisher for sensitive systems
Tool list, schemas, data path, and first read-only call
Explicit payment context before x402 or paid agent calls
The useful split is declared versus verified. A metadata label says what an agent claims. A live check says what the endpoint actually did when probed. Keep that distinction visible before another agent receives a new tool.
Recent S/A/B agents matched to this page. Inspect the profile before connecting a client or paying for a call.
MCP service candidate
Inspect its declared service, tool surface, and current score before connecting it to an agent client.
Inspect agentMCP service candidate
Inspect its declared service, tool surface, and current score before connecting it to an agent client.
Inspect agentSkills boundary candidate
Use the profile to separate workflow instructions from callable MCP, API, or web access.
Inspect agentMCP service candidate
Inspect its declared service, tool surface, and current score before connecting it to an agent client.
Inspect agent| Surface | Best at | Watch | Decision note |
|---|---|---|---|
| Docs lookup | Official docs MCP | Read-only workflow | Good first pattern for Codex integration work. |
| Issue or repo access | GitHub or Linear tools | Scoped writes | Use narrow auth and approval for changes. |
| Agent tools | ERC-8004 services | Identity and liveness | Inspect in The Spawn before adding to Codex. |
| Paid tools | x402 services | Amount and network checks | Never hide payment state from the user or agent. |
In a coding-agent workflow, MCP is useful when Codex needs a repeatable external interface instead of pasted context. A server can expose tools, resources, and prompts with typed inputs and outputs.
The useful mental model is not install more servers. It is give Codex one narrow interface that removes a repeated copy-paste or manual handoff.
Check job fit, publisher, transport, tool list, data path, approval model, maintenance, and first-call behavior. A good server has a small surface, clear schemas, boring errors, and a harmless first call.
A risky server has broad credentials, vague tools, hidden writes, no current documentation, or payment behavior that appears only after the model tries to use it.
The Spawn is for Codex users who want to discover and inspect MCP-capable onchain agents. It surfaces chain, token id, metadata URI, protocol declarations, liveness, quality, and x402 context.
The careful claim is discover and inspect, not guaranteed one-click execution. Listings differ: some expose MCP, some expose HTTP or A2A, some are paid, and some are only metadata.
No official OpenAI integration is claimed here. The Spawn helps inspect ERC-8004 agents that may expose MCP, HTTP, A2A, OASF, web, or x402 services.
A read-only official docs server is a safer first pattern than a server with writes, spending, or production access.
When the task needs a public agent or paid machine service and the endpoint, identity, and payment state have been inspected first.