The Humanoid Interop Standard (HIS) — draft

A proposed open standard for describing humanoid robot capabilities and letting LLM/VLA controllers discover and command any compliant robot without vendor-specific glue code — the same way MCP lets one LLM host talk to any tool server. This is a draft for discussion, not a ratified spec. Layer 0 — the capability manifest — is live, generated from the same cited data the HMX ranking already scores. Layer 3 — the tool-calling contract — has a draft schema and a working reference implementation in the site's repository, but is not connected to any real robot; no robot in this index implements a control protocol yet. Nothing below competes with ROS 2 or MCP; it's meant to be a thin certification/orchestration layer on top of both.

Live manifests across the diversity of the dataset

Live manifests need a free API key. The static examples above are open to everyone.

JSON API

The same cited dataset the manifests are generated from is readable directly as JSON with a free key: confirm your email and we send you the key and a working link. A key lasts 90 days and allows 1,000 requests per UTC day. Send it as the X-API-Key header or the api_key query parameter. Every endpoint returns exactly what the site itself renders, so a figure you disagree with can be checked against its source rather than argued about.

• Whole dataset: GET /api/robots — every robot with its scores and rank. Optional filters: robot_class, status, country, payload_min, speed_min, dof_min (e.g. /api/robots?robot_class=humanoid&payload_min=20); include=listed adds the robots listed but not ranked. There is no sort parameter — ranking order is the methodology's, not a query option.
• One robot: GET /api/robot/{slug} — the same record on its own.
• Evidence history: GET /api/v1/robots/{slug}/snapshots, and one snapshot at GET /api/v1/snapshots/{id}.
• Cited research export: GET /api/v1/research-export?slugs=…, optionally with as_of= an ISO timestamp.
• Catalogue changes: GET /api/v1/changes?slugs=… — what Watch shows.
• Finder results: GET /api/finder?task=… — the Finder's screening results.
• Scoring rules: GET /api/methodology — pillar weights, autonomy and status bases, and reference scales, so any HMX score in /api/robots is reproducible without scraping Methodology.
• Capability manifest: GET /robot/{slug}/manifest.json — the Layer 0 view of the same robot, described above.

Layer 3: the tool-calling contract (draft, not connected to a real robot)

Four JSON-RPC methods — initialize, tools/list, tools/call, tools/confirm — modeled on MCP's tool-discovery shape, plus a risk_tier on every tool that a generic tool protocol doesn't need but a robot control protocol can't skip: a bad MCP call corrupts a file, a bad robot call can move mass near a human. A call whose risk tier is at or above the robot's declared confirmation_required_above_tier returns pending_confirmation instead of executing — the orchestrator must make a second, explicit call to actually run it.

The reference implementation is a simulated Figure 02 tool server exposing four Tier-1 skills (speak, navigate_to, grasp, hand_off) with real gating behavior — every call is asserted end-to-end, not just documented. It lives in the repository, not on this server, because there's nothing real for it to control yet.

• Spec: standard/layer3-tool-contract.md
• Tool schema: tool-definition.schema.json
• Run it: python -m standard.reference.demo — prints every JSON-RPC message in the full initialize → tools/list → tools/call → tools/confirm flow

Certification levels (Layer 5 — schema exists, no test suite yet)

Every robot on this site is currently documented — no Layer 3 control protocol exists yet for any of them, so conformant and interop-proven are aspirational levels defined in the schema, not achieved by anything today. interop-proven stays aspirational for a second, independent reason: NIST has published the benchmark's four capability areas but not its task list or metrics, which are stated for release in summer 2026. See Benchmarks for what is and is not published.