launch

Launch Code OSS (VS Code from sources) into an isolated throwaway profile with unique debug ports so you can drive it with @playwright/cli AND attach a Node debugger via dap-cli in the same session. Use when working on VS Code itself and you want to interact with the running workbench, automate chat or UI flows, test UI features, take screenshots, set breakpoints in the renderer / extension host / main process, or combine UI driving with debugging.

Quality

86%

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

85%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.

Highly actionable, executable content with an unusually strong troubleshooting and validation story; almost nothing is generic knowledge Claude already has. The main weakness is structure: a 377-line monolithic body with no reference-file layer, plus a parallel Windows track that doubles length inline.

Suggestions

Move the Windows-specific material (PowerShell launch examples, ConvertFrom-Json capture, fnm PATH setup, Windows auth details) into a references/windows.md and keep a one-line pointer in the body.

Extract the troubleshooting catalog into references/troubleshooting.md, leaving the top 2-3 most common failures inline with pointers to the rest.

Move the 'Why a helper script' internals (NBSP normalization, async DOM update traps) next to monaco-paste.sh in a reference and keep only the usage contract inline, cutting the body well under 250 lines.

Dimension	Reasoning	Score
Conciseness	Nearly every line carries non-obvious, skill-specific knowledge Claude cannot already know (Monaco native-edit-context paste trap, NBSP normalization, cliDaemon session isolation, the ~103-byte unix socket limit), but the full Windows parallel track and some repeated rationale notes could be trimmed — minor over-explanation rather than efficient, lean prose.	4 / 5
Actionability	Fully executable, copy-paste-ready bash and PowerShell blocks with exact flags, the JSON output contract, helper exit-code semantics (0/1/2), and concrete per-symptom troubleshooting commands — covers the common cases completely.	5 / 5
Workflow Clarity	Clear sequence (prerequisites → launch → capture ports from JSON → attach → drive/verify → restart → cleanup) with explicit validation checkpoints (paste verify read-back with exit 1 on failure, tab-list target checks before interacting, non-zero exit plus log tail on launch failure, warning not to treat completed CLI commands as proof of entry) and feedback loops for recovery.	5 / 5
Progressive Disclosure	Scripts are correctly externalized (launch.sh, launch.ps1, monaco-paste.sh all exist and are well-signaled), but the 377-line SKILL.md inlines substantial content that clearly belongs in separate reference files — the entire Windows-specific guide, the Monaco-paste internals section, and the deep troubleshooting catalog — with no references/ layer at all.	3 / 5
	Total	17 / 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 description with explicit what/when structure, concrete named tools, and natural trigger phrases for a well-defined niche. The only deductions are the second-person voice (specificity penalty per rubric) and a few missing natural synonyms.

Suggestions

Rewrite in third person (e.g. "Launches Code OSS... so an agent can drive it") to avoid the rubric's second-person specificity penalty.

Add a few natural synonyms users might say, e.g. "Copilot chat automation", "Electron debugging", "UI automation".

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions ("Launch Code OSS... into an isolated throwaway profile with unique debug ports", "drive it with @playwright/cli", "attach a Node debugger via dap-cli", "take screenshots", "set breakpoints in the renderer / extension host / main process"), matching the comprehensive anchor 5, but the second-person phrasing ("so you can drive it", "when... you want to interact") triggers the rubric's 1-point specificity penalty.	4 / 5
Completeness	Explicitly answers both what (launch an isolated Code OSS instance with unique debug ports drivable via @playwright/cli and dap-cli) and when ("Use when working on VS Code itself and you want to interact with the running workbench, automate chat or UI flows..."), with concrete trigger phrases matching the top anchor.	5 / 5
Trigger Term Quality	Good natural-phrase coverage ("working on VS Code itself", "automate chat or UI flows", "test UI features", "take screenshots", "set breakpoints", "combine UI driving with debugging"), but a few natural terms a user might say are missing (e.g. "Copilot", "Electron", "UI automation"), so it sits between the good-coverage and comprehensive anchors.	4 / 5
Distinctiveness Conflict Risk	Clear niche (VS Code built from sources / Code OSS, @playwright/cli over CDP, dap-cli debug ports) with triggers unlikely to fire for any other skill; minimal conflict risk.	5 / 5
	Total	18 / 20 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: posit-dev/positron
Path: .agents/skills/launch/SKILL.md
Commit: b0258bc

Reviewed: 5 days 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.