The MCP setup guide walks through configuring Claude or any MCP-compatible client against a local server.
This isn't a mockup. Your browser fetches the live Traverse capability registry, picks a group of related capabilities, and chains them into a real, working pipeline -- the same kind of workflow you could compose today with traverse-cli. Nothing on this page is precomputed or faked.
The data is real: your browser fetches the live Traverse capability registry and a client-side heuristic -- not a hosted AI model -- groups related capabilities and orders them into a plausible pipeline, the shape of what an AI agent could compose over MCP. Nothing is precomputed; only the WASM execution afterward is simulated, not actually run. Full breakdown of what's real and what's not in how this is actually built, below.
How to read it: your Goal is the request that starts everything. The Traverse system -- the AI agent, its MCP connection, and the runtime -- discovers and validates a matching workflow along the dashed line. The runtime then chains together the right capabilities to do the actual work, producing a Result along the solid line. Watch each step light up as it runs, and click any capability to open its real, published contract in the registry.
+How to read it: your Goal is the request that starts everything. The Traverse system -- the AI agent, its MCP connection, and the runtime -- is the architecture a real agent would talk to, discovering and validating a matching workflow along the dashed line. The runtime then chains together the right capabilities to do the actual work, producing a Result along the solid line. Click any capability to open its real, published contract in the registry. The step-by-step animation shows execution order; it doesn't run the WASM.
\n Traverse is a contract-driven WASM runtime. Write business logic once with a machine-readable contract, run it natively or in the browser today, and let AI agents call it safely over MCP.\n
\n[capability]\nname = \"calculate-price\"\nversion = \"1.2.0\"\nwasm = \"pricing.wasm\"\n\n[governing_spec]\ntitle = \"Pricing Policy v3\"\nsource = \"docs/pricing-policy.md\"\n\n[input]\ntype = \"object\"\nproperties.sku = \"string\"\nproperties.quantity = \"integer\"\nproperties.region = \"string\"\nrequired = [\"sku\", \"quantity\"]\n\n[output]\ntype = \"object\"\nproperties.total = \"number\"\nproperties.currency = \"string\"\nrequired = [\"total\", \"currency\"]\n
\n Three steps from idea to a portable, contract-governed piece of business logic running in any environment.\n
\nWrite a TOML file with your capability name, the WASM binary path, and JSON Schema definitions for inputs and outputs. Link it to the governing spec document your team actually works from.
\nImplement the logic in Rust and compile to wasm32-wasi. The same binary runs on every placement target without recompilation.
Add the capability to the registry with traverse add. Call it from the CLI, from an AI agent via MCP, or from your app directly. Every call is validated and traced.
From contract validation to AI agent discovery, Traverse handles the entire lifecycle.
\nEvery capability call validates inputs against a JSON Schema precondition before execution. Postconditions validate the output. Bad data never reaches your logic.
\nEach capability runs in its own Wasmtime WASM instance with linear memory isolation. A capability cannot access host resources it was not explicitly granted.
\nExpose your entire capability registry to AI agents via the MCP stdio server. Claude, GPT-4, and any MCP-compatible agent can discover and call capabilities safely.
\nOne WASM binary runs on the native target today, and directly in the browser via the Web embedder SDK. Edge and cloud placement are on the roadmap, not yet shipped.
\nEvery execution produces a trace artifact: the input, the output, the contract version, and a timestamp. Audit any past run without re-executing anything.
\nEvery contract links to the governing spec document it implements. When the policy changes, you update the spec and bump the contract version. The paper trail is built in.
\nPricing rules, eligibility checks, compliance logic — any deterministic computation that needs to run reliably everywhere.
\nEncode complex pricing rules as a contract with a link to your pricing policy document. The same capability runs in your frontend, your backend, and inside AI agent pipelines with identical results.
\n Learn more →\nRun eligibility logic as a sandboxed WASM capability so an AI agent can check a user's eligibility for a product, service, or feature without risk of hallucinating the answer.
\n Learn more →\nCapture regulatory and compliance rules in versioned contracts with governing spec links. Every execution is traced. Auditors can replay any historical decision from the trace artifact.
\n Learn more →\nGive your AI agents a set of contract-governed capabilities to call. The agent cannot exceed the contract's boundaries. Postcondition checks block invalid outputs before they propagate.
\n Learn more →\nInstall the CLI, write your first contract, and run it in under five minutes.
\n\n Traverse is a contract-driven WASM runtime. Write business logic once with a machine-readable contract, run it natively or in the browser today, and let AI agents call it safely over MCP.\n
\n[capability]\nname = \"calculate-price\"\nversion = \"1.2.0\"\nwasm = \"pricing.wasm\"\n\n[governing_spec]\ntitle = \"Pricing Policy v3\"\nsource = \"docs/pricing-policy.md\"\n\n[input]\ntype = \"object\"\nproperties.sku = \"string\"\nproperties.quantity = \"integer\"\nproperties.region = \"string\"\nrequired = [\"sku\", \"quantity\"]\n\n[output]\ntype = \"object\"\nproperties.total = \"number\"\nproperties.currency = \"string\"\nrequired = [\"total\", \"currency\"]\n
\n Three steps from idea to a portable, contract-governed piece of business logic running in any environment.\n
\nWrite a TOML file with your capability name, the WASM binary path, and JSON Schema definitions for inputs and outputs. Link it to the governing spec document your team actually works from.
\nImplement the logic in Rust and compile to wasm32-wasi. The same binary runs on every placement target without recompilation.
Add the capability to the registry with traverse add. Call it from the CLI, from an AI agent via MCP, or from your app directly. Every call is validated and traced.
From contract validation to AI agent discovery, Traverse handles the entire lifecycle.
\nEvery capability call validates inputs against a JSON Schema precondition before execution. Postconditions validate the output. Bad data never reaches your logic.
\nEach capability runs in its own Wasmtime WASM instance with linear memory isolation. A capability cannot access host resources it was not explicitly granted.
\nExpose your entire capability registry to AI agents via the MCP stdio server. Claude, GPT-4, and any MCP-compatible agent can discover and call capabilities safely.
\nOne WASM binary runs on the native target today, and directly in the browser via the Web embedder SDK. Edge and cloud placement are on the roadmap, not yet shipped.
\nEvery execution produces a trace artifact: the input, the output, the contract version, and a timestamp. Audit any past run without re-executing anything.
\nEvery contract links to the governing spec document it implements. When the policy changes, you update the spec and bump the contract version. The paper trail is built in.
\nYour browser fetches the real capability registry and a client-side heuristic -- not a hosted AI model -- chains matching capabilities into a workflow. No fixtures, no precomputed data.
\n Watch the live demo →\nPricing rules, eligibility checks, compliance logic — any deterministic computation that needs to run reliably everywhere.
\nEncode complex pricing rules as a contract with a link to your pricing policy document. The same capability runs in your frontend, your backend, and inside AI agent pipelines with identical results.
\n Learn more →\nRun eligibility logic as a sandboxed WASM capability so an AI agent can check a user's eligibility for a product, service, or feature without risk of hallucinating the answer.
\n Learn more →\nCapture regulatory and compliance rules in versioned contracts with governing spec links. Every execution is traced. Auditors can replay any historical decision from the trace artifact.
\n Learn more →\nGive your AI agents a set of contract-governed capabilities to call. The agent cannot exceed the contract's boundaries. Postcondition checks block invalid outputs before they propagate.
\n Learn more →\nInstall the CLI, write your first contract, and run it in under five minutes.
\nCheck the registry before authoring a new capability — something close enough to reuse or extend might already exist.