Add robot profile: boston_dynamics/spot/boston_dynamics.spot.mujoco-pybullet-sim.v1

[EslaM-X]

🦾 Boston Dynamics Spot · Tier 1 Simulator Submission

Profile: boston_dynamics/spot/boston_dynamics.spot.mujoco-pybullet-sim.v1
Scope: Simulator-only · 8 paid skills · MuJoCo + PyBullet

A paid RoboPay action arriving on the Zenoh topic robot/tunnel/action
drives the official MuJoCo Spot model (mujoco_menagerie) through real,
policy-triggered skill episodes — joint-space control with body-weight
compensation, never a replayed animation or a built-in demo motion.

---

What's in the box

Path	Purpose
registry/vendors/boston_dynamics/spot/boston_dynamics.spot.mujoco-pybullet-sim.v1/	Full robot action profile (robot · skills · functions · payment policy · execution mapping · example envelope · skill-contract tests · validation report)
simulation/spot/	Skill controller, x402 payment gate, Zenoh action link, 4 test suites
simulation/pybullet/	Sim-to-sim validation against the kinematic URDF + machine-readable report
simulation/setup.sh	Pinned, reproducible fetch of the official Spot model assets

The 8 skills (all priced, all measurable)

wave · sit · stand · stop (safe stop) · bow · nod · turn_to_face · hold

Every skill is executed by a joint-space trajectory controller and reports
real simulator metrics — paw lift height, sit depth, torso pitch/yaw,
body height, achieved heading — then returns to the home stance
(|bodyZ − 0.434| < 0.02) so paid actions run back to back. stop is the
fail-safe skill: it halts motion and returns the robot to its stable home
pose (the wiki review checklist requires safe-stop behavior to be described).

Success criteria — cleared

• ✅ End-to-end paid flow — paid action → Zenoh → skill episode → correlated result on robot/tunnel/result
• ✅ Zenoh bridge — subscribes on robot/tunnel/action, republishes with actionId, responds on the documented result topic
• ✅ Real action requirement — 8 non-trivial embodied skills, visually demonstrable (screen recording: simulation/docs/spot.gif)
• ✅ Payment gate — unpaid ⇒ 402 + PAYMENT-REQUIRED; forged/expired receipts ⇒ 402; replayed idempotencyKey/txHash ⇒ 409; only status: success may settle
• ✅ Success & failure semantics — success + 6 deliberately testable failure paths (UNPAID, INVALID_PARAMS, UNKNOWN_SKILL, WRONG_ROBOT, DUPLICATE, tampered paramsHash)
• ✅ Reproducible — clean checkout + 4 commands, under 30 minutes, on Windows or ubuntu-latest (CI)
• ✅ Safe stop — stop skill halts motion and returns to the stable home stance (wiki review checklist)
• ✅ Code quality — no secrets, configurable topics/robotId/model path, JSON logs, 4 test suites
• ✅ Sim-to-sim — same joint configurations recomputed in MuJoCo and PyBullet agree to 0.06 cm across all salient poses (simulation/pybullet/sim2sim_report.json)

End-to-end flow

paid action (x402) → tunnel → Zenoh robot/tunnel/action
    → robopay_link.py → validate envelope + payment gate
    → joint-space Spot controller on mujoco_menagerie
    → metrics → result on robot/tunnel/result (correlated by actionId)

Evidence

• Screen recording (simulated action): simulation/docs/spot.gif (7 MB) — the 8 paid skills executed on the official MuJoCo Spot model
• Per-skill screenshots: simulation/docs/spot-shots/ (wave peak, sit crouch, bow, nod, turn, stance)
• Terminal logs (paid request → Zenoh → execution → result): simulation/docs/ci-logs.txt
• Architecture diagram: docs/images/flow.png
• Sim-to-sim report: simulation/pybullet/sim2sim_report.json
• Per-check JSON logs printed by every test suite (PASS/FAIL, nonzero exit on failure)
• CI: spot-simulation-tests workflow runs the full suite on ubuntu-latest — green on the fork (run 31497754516)

Reproduce in < 30 minutes

git clone https://github.com/fabricfoundation/RoboPay.git
cd RoboPay/simulation
./setup.sh                      # pinned official Spot model assets
cd spot
python3 test_spot_control.py    # every skill's physics actually happen
python3 test_payment_gate.py    # 402/409, no-settle-on-failure
python3 test_result_semantics.py# success/error semantics, replay, tampering
python3 test_link.py            # paid action → Zenoh → episode → result
cd ../pybullet
python3 test_sim2sim.py         # MuJoCo ⇄ PyBullet kinematics agreement

Known limitations (honest scope)

• Simulator-only submission — settlement is exercised end to end against the
  same local facilitator the robot link trusts; no on-chain settlement.
• The Go tunnel binary is not run on Windows; the Python payment gate
  reimplements the exact x402 decisions the tunnel's middleware makes before
  actuation, and the wire contract is exercised through peer-mode Zenoh exactly
  as the tunnel would publish it.

---

Built with care by EslaM-X 🇪🇬 — open to review feedback and happy to
iterate on any criterion.

[EslaM-X]

👋 Maintainers — ready for review

The full simulation suite is green on my fork:
spot-simulation-tests → success (Ubuntu, Python 3.12, 7 skills + payment
gate + result semantics + sim-to-sim, all passing).

The workflows on this PR are queued behind the first-time-contributor gate —
could you hit Approve and run? All three jobs are expected to pass:
Run Go Unit Tests, Run Go Lint, and spot-simulation-tests.

Review snapshot

• Scope: simulator-only, declared clearly in the profile and README.
• Payment safety: settle only on status: success; unpaid ⇒ 402, forged
  receipts ⇒ 402, replay ⇒ 409, tampered paramsHash ⇒ INVALID_PARAMS — all
  covered by test_payment_gate.py and test_result_semantics.py.
• Real action, not a demo: every skill is a joint-space trajectory driven
  by the controller on the official mujoco_menagerie model; metrics are
  measured, never scripted (screen recording: simulation/docs/spot.gif).
• Sim-to-sim: MuJoCo ⇄ PyBullet agree to 0.06 cm across all salient
  poses (simulation/pybullet/sim2sim_report.json).

Happy to address any review feedback.

— EslaM-X 🇪🇬

[EslaM-X]

✅ Update — 8th skill (safe stop) + terminal-log evidence

Quick follow-up before review:

• stop (safe stop) added as a priced skill — halts motion and returns
  the robot to the stable home stance on a short timeline. This covers the
  wiki review-checklist item "Emergency stop or safe stop behavior is
  described". Wired through the controller, skill catalog (JSON + YAML),
  execution mapping, payment policy and the acceptance test.
• Captured CI terminal logs added under simulation/docs/ci-logs.txt
  (paid action → Zenoh → episode → result), as required for simulator-only
  submissions.
• CI re-run green on the fork (run 31497754516) with the new skill.

Still needs Approve and run on this PR's workflows from a maintainer.

— EslaM-X 🇪🇬

[EslaM-X]

Hi @fabricfoundation — a quick update from my side 🙌

I've just added per-skill simulator screenshots (wave peak · sit crouch · bow · nod · turn · stance) next to the existing screen recording and the full terminal logs, so the paid skills are easy to eyeball without downloading anything.

Quick recap of where this submission stands:

• ✅ Full paid flow: action → Zenoh
  obot/tunnel/action\ → joint-space controller on the official MuJoCo Spot → correlated result on
  obot/tunnel/result\
• ✅ 8 priced skills, all measuring real simulator state (paw lift, sit depth, torso pitch/yaw, body height, achieved heading)
• ✅ Payment gate: \402\ unpaid · \409\ replay · forged/expired receipts rejected · no settle on failure
• ✅ 6 intentionally-testable failure paths, none of which actuate or settle
• ✅ Sim-to-sim MuJoCo ⇄ PyBullet agrees to 0.06 cm
• ✅ Reproducible from a clean checkout in < 30 min (Windows + ubuntu-latest CI, both green)

The honest scope is stated in the body: simulator-only, settlement exercised against the local facilitator — no on-chain settlement, no implied physical validation.

Open to any feedback — happy to iterate on whatever the reviewers want. 🦾

— EslaM-X 🇪🇬