Optional. Start with MCP setup: add the endpoint, finish the sign-in your client opens, then call proofable_context.
To load the same agent into an editor project:
Valid --apply values are cursor, claude, codex, hermes, openclaw, and opencode. VS Code uses --apply cursor. The command writes .proofable/mount.json and the host adapter. Hermes, OpenClaw, and OpenCode also get a pointer to that file at .hermes/proofable.json, .openclaw/proofable.json, or .opencode/proofable.json. Start a new session so the project picks it up.
The editor runs the tools. Proofable supplies identity, limits, and proofs. Cursor Cloud Agents connect to the same hosted MCP endpoint from the VMs Cursor runs.
From the agent card
Connect in copies the same command for the agent:
Run it in the project. If the host is not connected yet, run proofable setup first. Host files are written by this command, never by the browser.
When signed in, call proofable_agent_mount with agentId, agentWallet, or the identity proof ID in identityQHash.
Suggested order:
proofable_context
proofable_agent_mount when acting as a specific agent
proofable_verify_or_guide before sensitive actions; use proofable_proofs_check only for an explicit yes/no eligibility, gate, or access question
SDK
Use resolveRuntimeBundleFromMcp with your MCP transport, then applyRuntimeBundle(host, bundle, cwd) for project files. The proofable_agent_mount MCP tool and the proofable mount CLI return the same bundle shape.
Enforcement
Your client runs its own tools and enforces the mounted permissions in its own tool loop. Proofable enforces them directly for chat and jobs on Proofable. Agents compares each surface, and your first guarded action shows the check in code.
Last modified on October 25, 2026