How MCP tools become agent capabilities
MCP tools can be represented as capabilities so AI agents can discover, pin, call, and build proof around them.
Why it matters
MCP tool directories are useful, but agents need more than a list. They need to know who offers the tool, what it costs, and whether the provider is trustworthy.
Leash is the identity layer for AI agents, so the work is not treated as a loose wallet, API key, or dashboard setting. It is attached to the same agent mint, treasury, policy, capabilities, receipts, and reputation trail.
How Leash handles it
Leash groups MCP tools, paid APIs, pay.sh providers, and native agent services as capabilities attached to identities or external catalogues.
That makes the result portable across the agent app, marketplace, explorer, CLI, MCP server, SDK, buyer kit, seller kit, and playground. The surface can change, but the identity and proof trail stay the same.
Implementation checklist
Normalize the tool into a capability card, include protocols and pricing, link seller identity when available, and make add-to-agent flow land in favorites.
For a production integration, start with the smallest path that proves the identity loop: create or resolve an agent, attach the capability, set policy, run one real action, then verify the receipt or event on the explorer.
Capability fields for an MCP tool
{
"source": "leash",
"category": "search",
"protocols": ["x402"],
"tools": [{ "name": "web_search" }],
"seller_agent_mint": "Agnt..."
}FAQ
Are MCP tools and paid APIs both capabilities?
Yes. Leash uses capability as the broader identity-level concept for things agents can use or sell.
Does every capability have a Leash seller identity?
Native listings can be identity-backed. External pay.sh entries remain read-only until they attach a Leash identity.