rerun-chunk-processing

Core mechanics of the Rerun Chunk Processing API (rerun.experimental) — LazyChunkStream pipelines, Chunk, lenses (MutateLens/DeriveLens/Selector), RrdReader, writing optimized RRDs. Read BEFORE writing any ingestion/conversion/preprocessing code (convert an MCAP, build a recording from a dataset, preprocess an .rrd, port an old converter): it mandates reader+lens pipelines and steers away from hand-built chunks — no Chunk.from_columns for data a reader/lens can produce, no per-message rr.log, no manual pa.array assembly. Source-specific knowledge lives in the importer skills (rerun-mcap, rerun-urdf, rerun-parquet, rerun-mp4, rerun-lerobot); read rerun-data-model first to decide what the data should become.

Quality

84%

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

76%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.

Actionable and well-structured with strong executable examples, but the batch ingestion workflow lacks an explicit validation checkpoint and the single-file layout leaves the dense API reference inline.

Suggestions

Add an explicit validation/verification step to the ingestion workflow (e.g., reopen the RRD with RrdReader and assert chunk/entity counts) before declaring write_rrd complete, which would lift workflow_clarity above the batch-operation cap of 3.

Tighten or relocate the 'Why this order' rationale and the repeated anti-pattern explanations into a short reference file under references/ to improve conciseness and enable a cleaner progressive-disclosure split.

Split the Lens/Selector API mechanics and output-mode reference into a references/ file, keeping SKILL.md as the decision-rule overview with one-level-deep pointers.

Dimension	Reasoning	Score
Conciseness	Dense and assumes Claude's competence (no basic-concept padding), but explanatory prose such as the 'Why this order' paragraph and repeated anti-pattern rationales could be trimmed further.	4 / 5
Actionability	Multiple copy-paste-ready, executable code blocks (stream composition, MutateLens/DeriveLens, Chunk.from_columns, Selector) cover the common cases with specific API calls.	5 / 5
Workflow Clarity	The numbered 'Decision rule' gives a clear sequence, but this batch ingestion/conversion workflow has no explicit validation or verification checkpoint before write_rrd, so the destructive/batch cap holds at 3.	3 / 5
Progressive Disclosure	Well-organized section headers give clear navigation and external references are signaled, but no bundle files exist to offload the lengthy API mechanics into, so it is not a clean one-level-deep split.	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 highly specific, third-person description that clearly states both capabilities and use-conditions with concrete trigger phrases and file extensions. Minor trigger-term breadth keeps it just short of perfect.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions and API surfaces — 'LazyChunkStream pipelines, Chunk, lenses (MutateLens/DeriveLens/Selector), RrdReader, writing optimized RRDs' plus 'convert an MCAP, build a recording from a dataset, preprocess an .rrd, port an old converter' — comprehensive and specific.	5 / 5
Completeness	Explicitly answers both what ('Core mechanics of the Rerun Chunk Processing API...') and when ('Read BEFORE writing any ingestion/conversion/preprocessing code...') with concrete trigger phrases, so the completeness cap does not apply.	5 / 5
Trigger Term Quality	Natural trigger phrases like 'convert an MCAP', 'preprocess an .rrd', 'port an old converter' with file extensions (.rrd, MCAP) are present, but the terms skew technical and a few common variations are missing.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (Rerun Chunk Processing API) and explicitly distinguishes itself from the importer skills (rerun-mcap, rerun-urdf, etc.) and 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-chunk-processing/SKILL.md
Commit: 3efde05

Reviewed: 3 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.