[Newton] Add MuJoCo tendon support to Newton physics replication by hujc7 · Pull Request #5433 · isaac-sim/IsaacLab

hujc7 · 2026-04-28T23:41:12Z

Summary

Adds SchemaResolverMjc to proto-builder schema_resolvers so MjcTendon prims (e.g. Shadow Hand fixed tendons) are parsed correctly during Newton physics replication.
Fixes a root-cause bug ([BUG] MjcTendon joint-path resolution fails after proto + add_builder merge (silently drops all tendons) newton-physics/newton#2618) where including SchemaResolverMjc on the main builder caused two interlocking failures:
1. SchemaResolverMjc.validate_custom_attributes() raises unless SolverMuJoCo.register_custom_attributes() is called first.
2. register_custom_attributes registers MJC custom frequencies, triggering Newton's stage-wide traversal (independent of ignore_paths). On the main builder, this traversal finds MjcTendon prims under /World/envs/... and tries to resolve their joint paths against the main builder's empty joint_label, producing 128 "unknown joint path" warnings and silently dropping all tendons.
Fix: exclude SchemaResolverMjc from the main builder's schema_resolvers. The main builder loads only scene-level prims (ground, lights) with no MjcTendon prims. Proto builders include it with register_custom_attributes called first, resolving tendons correctly against each proto's populated joint_label. add_builder then propagates N×T entries across N environments — handled natively by SolverMuJoCo via tendon_world filtering.

Test plan

16 unit tests in test/cloner/test_mjc_tendon_cloner.py pass (no Isaac Sim required)
- Tests verify main builder has no MJC frequencies (no stage-wide traversal)
- Tests verify proto builder has MJC frequencies and tendon_world attribute
- Tests verify N×T entries after N add_builder calls is Newton's expected state
- Tests verify SchemaResolverMjc.validate_custom_attributes raises/passes appropriately
Shadow Hand environment with SolverMuJoCo no longer produces "unknown joint path" warnings at startup

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

This PR fixes a MuJoCo tendon duplication bug in Newton physics replication by correctly separating schema resolvers between the main builder and proto builders. The fix excludes SchemaResolverMjc from the main builder (which loads only scene-level prims) and includes it only in proto builders where register_custom_attributes is called first. The implementation is correct and well-documented, with comprehensive unit tests.

Architecture Impact

Self-contained within isaaclab_newton module. The changes affect only _build_newton_builder_from_mapping() which is called by newton_physics_replicate() and newton_visualizer_prebuild(). The fix aligns IsaacLab's cloner with Newton's expected N×T tendon entry design where SolverMuJoCo filters by tendon_world == template_world. No downstream callers need modification since the function signature is unchanged.

Implementation Verdict

Ship it — The fix is architecturally sound and correctly addresses the root cause.

Test Coverage

The PR adds 8 unit tests that cover the key invariants:

Main builder has no MJC custom frequencies (3 tests)
Proto builder has MJC custom frequencies registered (3 tests)
SchemaResolverMjc validation behavior (4 tests)
Tendon propagation with correct world indices (6 tests)

However, the tests use synthetic tendon injection rather than actually exercising the _build_newton_builder_from_mapping() function. This is acceptable given the Newton/USD dependencies would require integration tests, but there's no direct regression test for the actual fixed code path.

CI Status

No CI checks available yet — cannot verify tests pass in CI.

Findings

🔵 Improvement: source/isaaclab_newton/changelog.d/5433.rst:31-33 — Missing newline before final paragraph
The changelog file lacks a blank line before the final paragraph "The resulting N×T tendon entries..." which may cause RST rendering issues. The ^^^^^ section header style is correct but the final two lines run together with the preceding paragraph.

🔵 Improvement: source/isaaclab_newton/test/cloner/test_mjc_tendon_cloner.py:1-300 — Filename mismatch with PR description
The test file is named test_mjc_tendon_cloner.py but the PR description references test_tendon_deduplication.py. This is a documentation inconsistency, not a bug, but the PR description should be updated to match the actual filename.

🔵 Improvement: source/isaaclab_newton/test/cloner/test_mjc_tendon_cloner.py:97-98 — Accessing private Newton API
The test helper _inject_tendon_entries() accesses builder._custom_frequency_counts which is a private attribute (underscore prefix). While necessary for testing, this creates a coupling to Newton's internal implementation that may break if Newton refactors. Consider adding a comment noting this dependency.

# NOTE: Accessing private _custom_frequency_counts; may break if Newton refactors
builder._custom_frequency_counts[_TENDON_FREQ] = ...

🟡 Warning: source/isaaclab_newton/isaaclab_newton/cloner/newton_replicate.py:69-73 — Proto resolver list created once, instances shared across all protos

_proto_resolvers = [SchemaResolverMjc(), SchemaResolverNewton(), SchemaResolverPhysx()]

The resolver instances are created once and passed to every p.add_usd() call in the loop. If any resolver maintains mutable state that persists across calls, this could cause cross-contamination between protos. Verify Newton's resolvers are stateless, or move instantiation inside the loop:

for src_path in sources:
    p = NewtonManager.create_builder(up_axis=up_axis)
    solvers.SolverMuJoCo.register_custom_attributes(p)
    p.add_usd(
        stage,
        root_path=src_path,
        load_visual_shapes=True,
        skip_mesh_approximation=True,
        schema_resolvers=[SchemaResolverMjc(), SchemaResolverNewton(), SchemaResolverPhysx()],
    )

🔵 Improvement: source/isaaclab_newton/changelog.d/5433.rst — PR description mentions version bump to 0.5.16 not reflected in changed files
The PR description states "Bumps isaaclab_newton to 0.5.16" but no version file changes are included in the diff. Either the version bump should be added, or the PR description should be corrected.

🔵 Improvement: source/isaaclab_newton/test/cloner/test_mjc_tendon_cloner.py:233 — Test doesn't exercise actual fix code path
The tests verify Newton builder behavior in isolation but don't test that _build_newton_builder_from_mapping() correctly uses separate resolver lists. A minimal integration test (even if skipped in CI due to dependencies) that calls the fixed function would provide stronger regression coverage.

hujc7 · 2026-04-29T03:11:56Z

@greptileai Review

greptile-apps · 2026-04-29T03:14:49Z

Greptile Summary

This PR adds MuJoCo tendon support to Newton physics replication by calling SolverMuJoCo.register_custom_attributes on each proto builder (not the main builder), scoping custom-frequency prim filters to prevent cross-source contamination in heterogeneous clone plans, and extending _rename_builder_labels to rewrite string-typed custom-attribute columns (e.g. mujoco:tendon_label) alongside built-in label arrays. The fix correctly addresses the two-part root-cause bug from newton-physics/newton#2618 and is accompanied by a thorough unit-test suite covering all three logical parts (filter scoping, proto builder traversal, and N×T multi-world semantics).

Confidence Score: 5/5

Safe to merge — logic is sound, well-commented, and backed by 16 new unit tests; only minor P2 documentation and test-style suggestions remain.

No P0 or P1 findings. The previous thread concern about startswith path-prefix false-positives is already addressed in the current code with the trailing-slash guard. Remaining findings are documentation and defensive-coding style, both P2.

No files require special attention. The test helpers in test_rename_builder_labels.py have a minor inconsistency with production defensive coding for constraint_mimic attributes.

Important Files Changed

Filename	Overview
source/isaaclab_newton/isaaclab_newton/cloner/newton_replicate.py	Core logic change: adds _scope_custom_frequencies helper, moves register_custom_attributes to proto builders only, and extends _rename_builder_labels with a second pass for string custom-attribute columns — all well-guarded and thoroughly commented.
source/isaaclab_newton/test/cloner/test_mjc_tendon_cloner.py	New test file covering filter scoping (Part 2), USD tendon parsing (Part 1+2 integration), and N×T multi-world propagation (Part 3); _FakePrim stub is adequate for filter lambda tests but Newton's original filter may call methods beyond GetPath/GetTypeName.
source/isaaclab_newton/test/cloner/test_rename_builder_labels.py	New test file for _rename_builder_labels; _inject_builtins uses getattr without a default for constraint_mimic, which would raise AttributeError if Newton doesn't expose that attribute, while production code silently skips it.
source/isaaclab_newton/docs/CHANGELOG.rst	Adds CHANGELOG entries for 0.5.26 and 0.5.27; extension.toml jumps straight to 0.5.27, which appears intentional (both fixes bundled in one release).
source/isaaclab_newton/config/extension.toml	Version bump from 0.5.25 to 0.5.27, consistent with the two CHANGELOG entries added.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    A["newton_physics_replicate()"] --> B["_build_newton_builder_from_mapping()"]
    B --> C["Create main builder\n(no register_custom_attributes)"]
    C --> D["main_builder.add_usd(\n  ignore_paths=['/World/envs'] + sources\n)"]
    B --> E["For each src_path in sources"]
    E --> F["Create proto builder p"]
    F --> G["SolverMuJoCo.register_custom_attributes(p)\nregisters mujoco:* custom frequencies"]
    G --> H["_scope_custom_frequencies(p, src_path)\npatches usd_prim_filter with prefix+'/'\nfor cross-source isolation"]
    H --> I["p.add_usd(stage, root_path=src_path)\ntraversal scoped to src_path only"]
    I --> J["protos[src_path] = p"]
    J --> E
    B --> K["For each env col\nbegin_world / add_builder / end_world"]
    K --> L["N x T tendon entries in main builder\n(N envs x T tendons per proto)"]
    L --> M["Return builder"]
    M --> N["_rename_builder_labels()\nPass 1: built-in label arrays\nPass 2: string custom-attr columns\nwith references=world companion"]
    N --> O["NewtonManager.set_builder(builder)"]

_{Reviews (3): Last reviewed commit: "Apply ruff format to test_rename_builder..." | Re-trigger Greptile}

greptile-apps · 2026-04-29T03:14:53Z

+            continue
+        orig = freq.usd_prim_filter


Path-prefix false-positive in scope filter

startswith(_path) without a trailing / will match sibling USD paths that share a common string prefix. For example, if root_path="/World/src", a MjcTendon prim at /World/src2/tendon_coupling would incorrectly pass the filter, causing that prim to be resolved against the wrong proto's joint_label — exactly the zombie-tendon-entry problem the filter is meant to prevent.

The fix is to append "/" to the prefix check:

Suggested change

continue

orig = freq.usd_prim_filter

freq.usd_prim_filter = lambda prim, ctx, _orig=orig, _path=root_path: (

str(prim.GetPath()).startswith(_path + "/") and _orig(prim, ctx)

)

greptile-apps · 2026-04-29T03:14:54Z

+        """Plain builder has no MJC frequencies; _scope_custom_frequencies must not raise."""
+        b = _plain_builder()
+        _scope_custom_frequencies(b, self.ROOT_A)
+
+    def test_heterogeneous_plan_no_cross_contamination(self):
+        """Each proto only accumulates tendon entries from its own source subtree.
+
+        Simulates the heterogeneous scenario: two proto builders, each scoped to
+        a different source path.  After both receive tendon entries representing
+        their own prims (as the scoped traversal would inject), neither contains
+        entries from the other source.
+        """


Test gap: path-prefix ambiguity not covered

TestScopeMjcTendonFilters tests a path from ROOT_B being correctly rejected by a filter scoped to ROOT_A, but both roots are chosen so neither is a string prefix of the other (source_a / source_b). The latent bug in the production filter — where startswith without a trailing "/" also matches a sibling path like /World/envs/env_0/source_aa/tendon when scoped to /World/envs/env_0/source_a — is not exercised. A test like the following would expose the issue:

def test_filter_rejects_prim_at_sibling_with_shared_string_prefix(self): b = self._scoped_builder("/World/envs/env_0/source_a") prim = _FakePrim("/World/envs/env_0/source_aa/tendon") freq = b.custom_frequencies[_TENDON_FREQ] self.assertFalse(freq.usd_prim_filter(prim, {}))

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

Previous concern about shared resolver instances remains unaddressed. The code is identical to the last review — proto_resolvers is still instantiated once at line 97 and shared across all proto builders.

Implementation Verdict

Minor fixes needed — Move proto_resolvers = [SchemaResolverMjc(), SchemaResolverNewton(), SchemaResolverPhysx()] inside the for src_path in sources: loop (after line 103) to ensure each proto builder gets fresh resolver instances, or add a comment confirming Newton resolvers are stateless.

CI Status

Pre-commit is failing — this needs to be resolved before merge.

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

The previous concern about shared proto_resolvers instances remains unaddressed. The code at lines 97 and 113 is identical to the last review — proto_resolvers is still instantiated once outside the loop and reused across all proto builders.

Implementation Verdict

Minor fixes needed — Either move proto_resolvers = [SchemaResolverMjc(), SchemaResolverNewton(), SchemaResolverPhysx()] inside the for src_path in sources: loop (after line 103), or add a comment confirming Newton resolvers are stateless and safe to share.

CI Status

Pre-commit is still failing — this must be resolved before merge.

hujc7 · 2026-04-29T04:21:29Z

@greptileai Review

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

Previous concern about shared proto_resolvers has been addressed — a comment was added at line 97-99 confirming the resolvers are stateless and safe to share. Pre-commit is now passing.

Implementation Verdict

Ship it — The fix is correct, well-documented, and thoroughly tested.

Findings

No new issues. The implementation correctly isolates SchemaResolverMjc from the main builder and scopes custom frequencies to prevent cross-source contamination. Test coverage is comprehensive.

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

No new commits since the previous review — the HEAD is still 8a39a8f8. The implementation remains correct: SchemaResolverMjc is properly isolated from the main builder, _scope_custom_frequencies prevents cross-source contamination, and the test suite comprehensively validates all three parts of the fix.

Implementation Verdict

Ship it — Ready to merge pending CI completion.

isaaclab-review-bot

Follow-up review on commit 6cea8f71. No code changes since the previous review — the implementation remains correct and well-tested. The MuJoCo tendon support for Newton physics replication is ready to ship.

CI Status

Several CI jobs are failing but appear unrelated to this PR:

Check changelog fragments: failure — likely needs a changelog fragment file (standard PR process issue)
Installation Tests: failure — needs investigation but not caused by this PR's changes
Pre-commit passed ✅

Implementation Verdict

Ship it — pending CI infrastructure fixes and changelog fragment addition.

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

Follow-up review on commit 70302def. The changelog fragment has been added (jichuanh-newton-replicate-tendon-fix.rst), addressing the previous CI concern about missing changelog fragments. No functional code changes since the last review — the implementation remains correct and comprehensive.

CI Status

Check changelog fragments: success ✅ — Previous concern addressed
Check for Broken Links: failure ❌ — Unrelated to this PR (likely a docs infrastructure issue)
Most jobs still pending; pre-commit passed ✅

Implementation Verdict

Ship it — The changelog is well-written with clear "Added" and "Fixed" sections that accurately describe the changes. Ready to merge once CI passes.

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

Follow-up review on commit d8383ad0. No code changes since the last review — this appears to be a CI re-run or merge commit. The implementation remains correct and comprehensive.

CI Status

Check changelog fragments: success ✅
pre-commit: success ✅
Build Wheel: success ✅
Check for Broken Links: failure ❌ — Unrelated to this PR (docs infrastructure issue)
Most other jobs still pending

Implementation Verdict

Ship it — All previous concerns have been addressed. The changelog is present and accurate, the implementation is sound, and tests are thorough. Ready to merge once remaining CI jobs pass.

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

Follow-up review on commit 54d0eca5. No substantive code changes since the previous review — the implementation remains correct and well-tested. All previous concerns were addressed in earlier iterations.

CI Status

All jobs still pending. Pre-commit and Build Wheel passed previously; waiting on isaaclab_newton tests which are the relevant jobs for this PR.

Implementation Verdict

Ship it — Ready to merge once CI completes successfully.

The Newton 1.2 release includes the upstream tendon-scoping fix (newton-physics/newton#2659), which scopes ``parse_usd``'s custom- frequency walk to ``root_path`` natively. Bumping the Newton pin removes the cross-source ``MjcTendon`` contamination IsaacLab previously had to work around at the framework layer. Pin bumps: * ``source/isaaclab/setup.py``: ``warp-lang==1.12.0`` → ``>=1.13.0``, ``mujoco==3.6.0`` → ``==3.8.0``, ``mujoco-warp==3.6.0`` → ``==3.8.0.1``. * ``source/isaaclab_newton/setup.py``: same mujoco / mujoco-warp bumps; Newton pin → ``v1.2.0rc1``. * ``source/isaaclab_visualizers/setup.py``: 3 × Newton pin → ``v1.2.0rc1``. Code adapts: * ``source/isaaclab_newton/isaaclab_newton/physics/newton_manager.py`` and ``source/isaaclab_ov/isaaclab_ov/renderers/ovrtx_renderer_kernels.py``: ``wp.math.transform_to_matrix`` → ``wp.transform_to_matrix``. The ``wp.math`` namespace was removed in ``warp-lang`` 1.13. Per-env label rename: * ``_rename_builder_labels`` now walks string-typed custom-attribute columns whose frequency declares a ``references="world"`` companion (e.g. ``mujoco:tendon_label``), rewriting the per-row source-path prefix to the destination world root. ``constraint_mimic`` joins the built-in label pass. The prefix match uses a path-separator boundary so source paths that share a string prefix (``/Sources/protoA`` vs ``/Sources/protoAB``) do not cross- contaminate. Newton's ``add_builder`` copies labels verbatim and tracks env identity in ``*_world`` int columns; the IL rename is the structural translation between Newton's labeling model and IL's per-env USD-path model. Both passes (built-in label arrays and string custom-attribute columns) remain necessary after the Newton bump. Adds ``test_rename_builder_labels.py`` with 10 cases covering the two passes, multi-source prefix-overlap regression, sparse env ids, multi-frequency, multiple string columns at one frequency, and empty-values pass-through.

isaaclab-review-bot

🤖 Isaac Lab Review Bot

Summary

Follow-up review on commit fce4ebdf. The diff shows the same implementation as the previous review - no new changes since the last "Ship it" verdict. The dependency bumps (Newton v1.2.0rc1, warp-lang>=1.13.0, mujoco 3.8.0), the _rename_builder_labels fix with boundary-terminated prefix matching, the wp.transform_to_matrix API update, and the comprehensive test suite are all unchanged.

CI Status

Most jobs still pending. Pre-commit passed ✅. The "Check changelog fragments" failure needs attention - verify the changelog fragment format meets project requirements (the .rst extension and location look correct, but the CI may have specific validation rules).

Implementation Verdict

Ship it — Code is ready. Address the changelog fragment CI failure (likely a formatting or validation issue) and merge once remaining CI jobs pass.

hujc7 requested review from ClemensSchwarke, Mayankm96, Toni-SM, hhansen-bdai, jtigue-bdai, kellyguo11, masoudmoghani, matthewtrepte, ooctipus, ossamaAhmed, pascal-roth, peterd-NV and xyao-nv as code owners April 28, 2026 23:41

github-actions Bot added bug Something isn't working documentation Improvements or additions to documentation asset New asset feature or request isaac-sim Related to Isaac Sim team isaac-mimic Related to Isaac Mimic team infrastructure labels Apr 28, 2026

hujc7 marked this pull request as draft April 29, 2026 01:10

hujc7 pushed a commit to hujc7/IsaacLab that referenced this pull request Apr 29, 2026

Add changelog fragment for PR isaac-sim#5433

500ab8a

hujc7 force-pushed the jichuanh/newton-replicate-tendon-fix branch from ea456bc to 500ab8a Compare April 29, 2026 01:11

hujc7 changed the base branch from main to develop April 29, 2026 01:12

github-actions Bot added the isaac-lab Related to Isaac Lab team label Apr 29, 2026

hujc7 force-pushed the jichuanh/newton-replicate-tendon-fix branch from 500ab8a to 99f7d7b Compare April 29, 2026 01:40

hujc7 changed the title ~~Fix MuJoCo tendon duplication in Newton physics replication~~ Add MuJoCo tendon support to Newton physics replication Apr 29, 2026

isaaclab-review-bot Bot reviewed Apr 29, 2026

View reviewed changes

hujc7 changed the title ~~Add MuJoCo tendon support to Newton physics replication~~ [Newton] Fix MuJoCo tendon duplication in Newton physics replication Apr 29, 2026

hujc7 changed the title ~~[Newton] Fix MuJoCo tendon duplication in Newton physics replication~~ Add MuJoCo tendon support to Newton physics replication Apr 29, 2026

hujc7 changed the title ~~Add MuJoCo tendon support to Newton physics replication~~ [Newton] Add MuJoCo tendon support to Newton physics replication Apr 29, 2026

greptile-apps Bot reviewed Apr 29, 2026

View reviewed changes

isaaclab-review-bot Bot reviewed Apr 29, 2026

View reviewed changes

hujc7 force-pushed the jichuanh/newton-replicate-tendon-fix branch from a6a2a7c to 8a39a8f Compare April 29, 2026 04:32

isaaclab-review-bot Bot reviewed Apr 29, 2026

View reviewed changes

hujc7 mentioned this pull request Apr 29, 2026

[WIP][Newton] Shadow-Hand-Over MAPPO: raise actuator stiffness so Newton learns the catch (depends on #5433) #5437

Draft

6 tasks

isaaclab-review-bot Bot reviewed May 1, 2026

View reviewed changes

hujc7 marked this pull request as ready for review May 1, 2026 09:16

isaaclab-review-bot Bot reviewed May 1, 2026

View reviewed changes

kellyguo11 added this to Isaac Lab May 1, 2026

kellyguo11 moved this to In progress in Isaac Lab May 1, 2026

isaaclab-review-bot Bot reviewed May 1, 2026

View reviewed changes

hujc7 force-pushed the jichuanh/newton-replicate-tendon-fix branch from 80ed8df to 6cea8f7 Compare May 2, 2026 18:55

isaaclab-review-bot Bot reviewed May 2, 2026

View reviewed changes

hujc7 force-pushed the jichuanh/newton-replicate-tendon-fix branch from 6cea8f7 to 70302de Compare May 2, 2026 19:01

isaaclab-review-bot Bot reviewed May 2, 2026

View reviewed changes

hujc7 force-pushed the jichuanh/newton-replicate-tendon-fix branch from 70302de to d8383ad Compare May 2, 2026 19:11

isaaclab-review-bot Bot reviewed May 2, 2026

View reviewed changes

hujc7 force-pushed the jichuanh/newton-replicate-tendon-fix branch from d8383ad to 54d0eca Compare May 5, 2026 00:35

isaaclab-review-bot Bot reviewed May 5, 2026

View reviewed changes

hujc7 force-pushed the jichuanh/newton-replicate-tendon-fix branch from 54d0eca to fce4ebd Compare May 5, 2026 16:41

isaaclab-review-bot Bot reviewed May 5, 2026

View reviewed changes

Conversation

hujc7 commented Apr 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Summary

Test plan

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

🤖 Isaac Lab Review Bot

Summary

Architecture Impact

Implementation Verdict

Test Coverage

CI Status

Findings

Uh oh!

hujc7 commented Apr 29, 2026

Uh oh!

greptile-apps Bot commented Apr 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Greptile Summary

Confidence Score: 5/5

Important Files Changed

Flowchart

Uh oh!

greptile-apps Bot Apr 29, 2026

Choose a reason for hiding this comment

Uh oh!

greptile-apps Bot Apr 29, 2026

Choose a reason for hiding this comment

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

🤖 Isaac Lab Review Bot

Summary

Implementation Verdict

CI Status

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

🤖 Isaac Lab Review Bot

Summary

Implementation Verdict

CI Status

Uh oh!

hujc7 commented Apr 29, 2026

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

🤖 Isaac Lab Review Bot

Summary

Implementation Verdict

Findings

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

🤖 Isaac Lab Review Bot

Summary

Implementation Verdict

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

🤖 Isaac Lab Review Bot

Summary

Implementation Verdict

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

🤖 Isaac Lab Review Bot

Summary

Implementation Verdict

CI Status

Uh oh!

isaaclab-review-bot Bot left a comment

Choose a reason for hiding this comment

🤖 Isaac Lab Review Bot

Summary

Implementation Verdict

CI Status

Uh oh!

hujc7 commented Apr 28, 2026 •

edited

Loading

greptile-apps Bot commented Apr 29, 2026 •

edited

Loading