run-evals

do e2e tests, run e2e, validate feature, prove it works, PR proof, frame proof, pnpm evals. Launches OpenWork on Daytona or local Electron and runs the coded eval flows via CDP. Launch + run mechanics; the proof loop itself is the fraimz skill.

Quality

82%

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

100%

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

The body is a lean, actionable, well-sequenced runbook with explicit validation checkpoints and one-level-deep references, scoring at the top of the rubric across all dimensions. It respects token budget while giving copy-paste-ready commands and clear fallback/teardown handling.

Dimension	Reasoning	Score
Conciseness	The body is lean and command-heavy, assumes Claude knows what CDP/Electron/VNC are, and never pads with concept explanations — every token earns its place, matching the score-3 anchor.	3 / 3
Actionability	It provides concrete, copy-paste-ready commands with arguments (e.g. 'bash .devcontainer/test-on-daytona.sh <branch-or-commit> --artifacts-volume' and 'pnpm evals --flow <flow-id> --cdp-url ...') plus an executable CDP verification call, matching the score-3 anchor.	3 / 3
Workflow Clarity	The flow is explicitly sequenced (prereqs -> launch -> verify CDP -> run flows -> teardown) with validation checkpoints ('Verify the endpoint before running flows', the real success marker ws://127.0.0.1:9825) and an error-recovery path (inspect /tmp/electron.log), matching the score-3 anchor.	3 / 3
Progressive Disclosure	No bundle files are present, so this scores against structure: the body is self-contained, well-sectioned, and uses one-level-deep pointers ('see the fraimz skill and evals/README.md', 'Details: daytona-recording-artifacts') that are clearly signaled, matching the score-3 anchor for a self-contained skill.	3 / 3
	Total	12 / 12 Passed

Description

65%

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

The description is specific and trigger-keyword-rich for its domain, but is front-loaded with a raw keyword dump and lacks an explicit 'Use when...' clause, capping completeness. It distinguishes itself from a sibling skill only partially due to shared proof terminology.

Suggestions

Replace the leading keyword dump ('do e2e tests, run e2e, validate feature, prove it works, PR proof, frame proof, pnpm evals') with a clean list of concrete capabilities.

Add an explicit 'Use when...' clause stating when Claude should load this skill, e.g. 'Use when the user asks to run coded evals / prove a PR works against a live OpenWork app'.

Sharpen the boundary with the fraimz skill by dropping shared 'PR proof'/'frame proof' triggers so the description does not fire for the wrong skill.

Dimension	Reasoning	Score
Specificity	Names a concrete domain and actions ('Launches OpenWork on Daytona or local Electron and runs the coded eval flows via CDP'), but the leading 'do e2e tests, run e2e, validate feature...' is a keyword dump rather than a clean enumerated list of capabilities, so it stops at score 2 rather than 3.	2 / 3
Completeness	It clearly states what the skill does but has no explicit 'Use when...' trigger clause, so per the judging guideline the 'when' is only implied and completeness caps at 2.	2 / 3
Trigger Term Quality	Good coverage of natural phrases a user would say ('run e2e', 'validate feature', 'prove it works', 'pnpm evals'), matching the score-3 anchor despite some project jargon like 'frame proof' and 'PR proof'.	3 / 3
Distinctiveness Conflict Risk	It carves out a launch+run niche versus the fraimz skill, but shares overlapping proof jargon ('PR proof', 'frame proof') with that sibling skill, so it could still overlap with similar skills.	2 / 3
	Total	9 / 12 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: different-ai/openwork
Path: .opencode/skills/run-evals/SKILL.md
Commit: 1345d51

Reviewed: about 15 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.