---
title: "W&B EvalTable Previews: review, setup & limitations · Undominated.ai"
canonical: https://undominated.ai/skills/wandb-wandb-eval-tables/
description: "Converts existing W&B Table artifacts into new EvalTable previews without overwriting source runs, with stated size caps."
---

# W&B EvalTable Previews: review, setup & limitations · Undominated.ai

> Converts existing W&B Table artifacts into new EvalTable previews without overwriting source runs, with stated size caps.

[← Explore all skills](/skills/)

MODEL EVALUATION / Weights & Biases

# W&B EvalTable Previews

Converts existing W&B Table artifacts into new EvalTable previews without overwriting source runs, with stated size caps.

 [See setup guidance ↓](#setup)[Original source ↗](https://github.com/wandb/skills/tree/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/skills/wandb-eval-tables)

SOURCE REVIEW

 Reviewed 2026-10-13
 Evidence 3 linked sources
 Publisher Weights & Biases
 Licence [Apache-2.0 ↗](https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/LICENSE)
 Revision 4efbb9fd82e8
 [Read what was—and wasn’t—checked ↓](#review)

“Convert W&B Table artifacts into non-destructive EvalTable previews with scan-first planning, typed input/output/score columns, bounded batches, verification, and safe removal. Use when a coding agent needs to create, inspect, compare, verify, or remove W&B EvalTable previews.”

 [Weights & Biases · upstream description ↗](https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/skills/wandb-eval-tables/SKILL.md) Our analysis follows below.

01 / THE REASONING

## Why this made the selection.

 - The workflow runs a read-only scan first, previews one table key per invocation, and reports success only when verify-preview returns verified: true.
- The helper caps tables at 10,000 rows and 100 columns and refuses more than 50 eligible artifacts per key; the file says to surface every cap and never imply a truncated preview is complete.

### A good fit for

 - Turning logged prediction tables into EvalTable previews
- Removing helper-created previews after a dry run

### Weigh up before choosing

 - Conversion relies on scripts/table_artifact_to_eval_table.py and references/EVAL_TABLES.md, which were not opened. preview creates W&B runs, EvalTables and Weave evaluations.
- Removal deletes Weave evaluations and preview runs; the file requires delete-preview --dry-run and confirmation first. The README marks the skill experimental.

02 / THE REVIEW RECORD

## What we actually inspected.

Source review has boundaries. A clear record is more useful than a “safe” badge.

### Material inspected

 - skills/wandb-eval-tables/SKILL.md (environment, required workflow, safety)
- README.md (requirements, skill table, installation)
- LICENSE Apache-2.0 text

### Our findings

 - Input tuples must uniquely identify rows, score columns must be numeric or boolean, and ambiguous tables stay untyped.
- Downloads go to the operating system's temporary directory, per the file.

### Not established by this review

 - The installer was not run, and the skill was not executed against a project or service. No W&B project was contacted.
- references/EVAL_TABLES.md and the two helper scripts were not opened.

The review applies to the material and revision named here. A newer upstream release can change its behavior.

03 / PUT IT TO WORK

## Add a skill to your workflow.

[Upstream setup instructions ↗](https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/README.md)

DOCUMENTED COMMAND

 npx skills add wandb/skills Copy command ↗

Run in your terminal to choose your AI tools, project or global scope, and installation method. This short command follows the current upstream version; the review covers the revision shown above.

 - The README documents npx skills add wandb/skills and selecting one skill with --skill <name>; the skill table names this skill wandb-eval-tables.
- The command does not pin revision 4efbb9fd82e8076cbe5d2332167b19f2bba90d85. Compare the installed SKILL.md, reference and helper scripts with that revision.

### Before you start

 - Node.js and npm for the documented Skills installer
- uv and a W&B SDK that provides wandb.EvalTable (the README states wandb >=0.28.1)

### Compatibility

Agent Skills-compatible hosts · Claude Code · Codex · W&B SDK with wandb.EvalTable and Weave

### Implementation

Python

### Host-mediated W&B project reads and writes

 - scan and preview --dry-run are read-only, per the file.
- preview creates runs, EvalTables and Weave evaluations in the target project; delete-preview removes them.
- Credentials and default scope may come from WANDB_API_KEY, WANDB_ENTITY and WANDB_PROJECT.

### Cost model

Source is Apache-2.0. W&B storage and compute can incur charges. Those charges were not measured.

04 / FOLLOW THE EVIDENCE

## The source trail.

Our notes are separate from the original resource. Check upstream before adopting a new version.

 - [Skill definition ↗](https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/skills/wandb-eval-tables/SKILL.md) Checked 2026-10-13 https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/skills/wandb-eval-tables/SKILL.md Supports: summary, upstreamDescription, whySelected, limitations, access, review
- [Install instructions and skill index ↗](https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/README.md) Checked 2026-10-13 https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/README.md Supports: install, compatibility
- [Repository licence ↗](https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/LICENSE) Checked 2026-10-13 https://github.com/wandb/skills/blob/4efbb9fd82e8076cbe5d2332167b19f2bba90d85/LICENSE Supports: license

KEEP COMPARING

## Other approaches to consider.

Related by category or shared topics. These are alternatives to inspect, not a measured quality order.

 [### Agent Platform Eval Flywheel ↗ Evaluates and iterates on Google Cloud Agent Platform models and agents, with a confirmation card before every billed evaluation run.](/skills/google-agent-platform-eval-flywheel/)[### Arize Evaluator ↗ Creates and runs Arize LLM-as-judge and code evaluators through the ax CLI, with a rule against fabricating scores.](/skills/arize-ai-arize-evaluator/)[### LangSmith Evaluator ↗ Builds LangSmith offline and online evaluators, with rules for name-based deletes and judge model configuration.](/skills/langchain-ai-langsmith-evaluator/)

 [AI Tools ↗](/tools/)[Skills ↗](/skills/)[Agents ↗](/agents/)[MCP Servers ↗](/mcp-servers/)[Workflows ↗](/workflows/)

## Continue your investigation

 - [Choose a toolkit](/tools/)
- [Inspect execution hosts](/agents/)
- [Check tool access](/mcp-servers/)
- [Check project dependencies](/check-repo/)
