rerun-data-model

How raw multimodal robot data maps onto the Rerun data model. Read FIRST, before modeling or converting a dataset — and whenever you are about to convert/ingest/preprocess robot data into an .rrd or build a Rerun recording, even if not asked for the data model. Resolves the entity-vs-component, property-vs-component-vs-layer, and static-vs-temporal decisions and routes to the mechanism (do it with readers and lenses, not hand-built chunks or per-message rr.log): rerun-chunk-processing and the importer skills rerun-mcap, rerun-urdf, rerun-parquet, rerun-mp4, rerun-lerobot.

Quality

78%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/rerun-data-model/SKILL.md

Quality

Content

68%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 well-structured, lean decision skill that routes mechanism details to siblings and gives concrete modeling rules plus a mapping-table artifact. Its main weakness is the absence of an explicit validation/verification checkpoint for the batch ingestion workflow it governs.

Suggestions

Add an explicit validation step after conversion — e.g. 'After ingest, open the .rrd and verify one entity per archetype renders on the expected timeline before proceeding' — to lift workflow_clarity above the batch-operation cap of 3.

Include a short fix→retry loop for the common failure modes already in the Gotchas section (e.g. layer not attaching → check recording_id match → re-emit), turning the gotchas into a feedback workflow.

Convert the sibling-skill pointers into explicit 'see <skill> for X' callouts at each decision branch so navigation is unambiguous rather than only listed up front.

Dimension	Reasoning	Score
Conciseness	Dense and assumes Claude's competence — no basic-concept padding, every section carries decision content — with only minor instances that could be trimmed, fitting 'efficient; minor instances of over-explanation' rather than the maximally lean 5.	4 / 5
Actionability	Gives concrete executable guidance for an instruction skill — 'New entity if it has its own spatial frame', 'Use AnyValues for non-standard fields', '[0][0] (0-based DataFrame, 1-based SQL)' and a fill-in mapping table — with minor gaps, sitting above the midpoint.	4 / 5
Workflow Clarity	A sequence exists (pick mechanism → model → decisions → mapping table) but dataset ingestion is a batch operation with no explicit validate/verify-record checkpoint or fix→retry feedback loop, which caps workflow clarity at 3 per the batch-operations guideline.	3 / 5
Progressive Disclosure	Clear overview ('This skill is just the decisions') with well-signaled one-level-deep routing to sibling skills and rerun.io docs, content appropriately split; it stops short of 5 because references are skill-name pointers rather than navigable file links and there are no local reference files.	4 / 5
	Total	15 / 20 Passed

Description

87%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 strong, third-person description that clearly states a distinct niche, concrete actions, and explicit when-to-use triggers with file extensions. It loses only minor points on trigger-term breadth and specificity coverage, not on completeness or distinctiveness.

Dimension	Reasoning	Score
Specificity	Names several concrete actions in a specific domain — 'maps onto the Rerun data model', 'Resolves the entity-vs-component...decisions', 'routes to the mechanism' — with only minor coverage gaps, fitting the 'lists several specific actions' anchor rather than the fully comprehensive 5.	4 / 5
Completeness	Explicitly answers both what (maps data, resolves decisions, routes to mechanism) and when ('Read FIRST, before modeling or converting a dataset — and whenever you are about to convert/ingest/preprocess robot data into an .rrd... even if not asked for the data model') with concrete trigger phrases, matching the top anchor.	5 / 5
Trigger Term Quality	Good keyword coverage with synonyms and a file extension — 'convert/ingest/preprocess robot data into an .rrd', 'build a Rerun recording', 'multimodal robot data' — but a few natural phrasings a user might say are absent, so it sits above the midpoint rather than at the comprehensive 5.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (Rerun data-model decisions for robot data) with distinct triggers and explicit routing to named sibling skills, giving minimal conflict risk per the top anchor.	5 / 5
	Total	18 / 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-data-model/SKILL.md
Commit: 3efde05

Reviewed: 2 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.