rerun-urdf

Drive the Rerun URDF API (rerun.urdf.UrdfTree) to ingest a URDF as a Transform3D layer on a robot recording. Read when logging a robot model, running forward kinematics from joint states, composing a fixed chain for sensor extrinsics, or when the transform tree will not connect from the data alone. Builds on rerun-chunk-processing (stream/lens mechanics) and rerun-data-model (entity paths, timeline, base-vs-layer).

Quality

83%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A dense, actionable body that defers generic stream/lens mechanics to sibling skills and concentrates on the URDF-specific judgment calls, with executable code and a validation-backed forest-resolution checklist. The main improvements are tightening the longer prose sections and considering a split reference file for the full pipeline example.

Suggestions

Tighten the 'How transforms compose' and 'Resolving the transform forest' prose — the six steps could be condensed into a shorter checklist with the graph-check validation called out as the key checkpoint.

Consider moving the full 'Minimal shape' pipeline and the calibration Chunk.from_columns edge into a references/ file, keeping SKILL.md as an overview that links to it, to improve progressive disclosure now that the file is ~250 lines.

Add an inline validation checkpoint in the 'Minimal shape' pipeline itself (e.g., 'confirm the joint-state stream yields rows and the forest has one root before write_rrd') so the workflow's validation is visible end-to-end rather than only in earlier sections.

Dimension	Reasoning	Score
Conciseness	Dense and assumes competence (no padding about what a URDF or quaternion is), but the multi-step 'Resolving the transform forest' and 'How transforms compose' prose could be tightened without losing the judgment-call guidance.	4 / 5
Actionability	Concrete, mostly copy-paste-ready code throughout (UrdfTree.from_file_path, compute_joint_transform_batches, a full two-lens 'Minimal shape' pipeline, a calibration Chunk.from_columns edge), with the few placeholders explicitly flagged as user-supplied data wiring.	4 / 5
Workflow Clarity	Clear two-step pipeline (stream static model -> solve FK -> merge -> write) plus a 6-step forest-resolution checklist with a pre-write graph check and a 'fail loudly or log identity + warning' feedback loop; validation is present but distributed across sections rather than inline in the final pipeline.	4 / 5
Progressive Disclosure	Well-organized into clear section headers with one-level-deep, clearly signaled references (two example URLs and the sibling skills rerun-data-model / rerun-chunk-processing), but all ~250 lines live in one file with no split reference bundle, leaving minor organization gaps.	4 / 5
	Total	16 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A precise, third-person description that names concrete API-driven actions and gives explicit 'Read when...' trigger guidance covering the skill's main scenarios. It is well-distinguished from sibling Rerun skills; the only gap is the absence of a '.urdf' file-extension trigger term.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'ingest a URDF as a Transform3D layer', 'logging a robot model', 'running forward kinematics from joint states', 'composing a fixed chain for sensor extrinsics', resolving an unconnected transform tree — giving comprehensive coverage of the skill's use cases.	5 / 5
Completeness	Explicitly answers both 'what' (drive rerun.urdf.UrdfTree to ingest a URDF as a Transform3D layer) and 'when' via a concrete 'Read when...' clause enumerating four trigger scenarios.	5 / 5
Trigger Term Quality	Strong natural terms ('URDF', 'robot model', 'forward kinematics', 'joint states', 'transform tree') but misses the '.urdf' file extension and a few synonyms, so it sits just below the comprehensive-with-extensions anchor.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (Rerun URDF API, Transform3D layer, forward kinematics) and explicitly distinguishes itself from sibling skills it builds on ('rerun-chunk-processing', 'rerun-data-model'), minimizing conflict risk.	5 / 5
	Total	19 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: rerun-io/rerun
Path: skills/rerun-urdf/SKILL.md
Commit: f30766b

Reviewed: ~7 hours ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.