Visual Studio Code is a code editor that supports editing, navigating, understanding, debugging, and extending software projects. Developers use it for the edit-build-debug cycle, and the catalogue add-ons provide skills, instructions, and agents for working within the editor.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add microsoft/vscode --skill record-videogit clone --depth 1 https://github.com/microsoft/vscodeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/microsoft/vscode/record-video)<a href="https://agentmods.dev/skills/microsoft/vscode/record-video"><img src="https://agentmods.dev/badge/skills/microsoft/vscode/record-video/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/microsoft/vscode/record-video"><img src="https://agentmods.dev/badge/skills/microsoft/vscode/record-video.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00061 | $0.00818 |
| Opus 5.5 | $0.00024 | $0.00327 |
| Sonnet 5.5 | $0.00012 | $0.00164 |
| Haiku 4.5 | $0.00006 | $0.00082 |
Grade A, and why
record-video scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Record video
Record a short .webm video of the real outcome with Playwright's
page.screencast, and return it with the script that produced it.
Use whichever Playwright page is available: the Integrated Browser's
runPlaywrightCode tool, or a Playwright session another skill attaches to an
app (for example @playwright/cli run-code). If a skill exists for launching
the app under test, follow it to get the page.
Rules
- Show real behavior. Never fake a result or hide a failed assertion.
- Don't bypass sign-in, CAPTCHAs, or paywalls; don't record secrets or private data; ask before side effects on real services.
- Keep it under a minute, in a few clear chapters.
Prepare
- Get a Playwright page for the app and make sure it is visible
(
document.visibilityState === 'visible'); ask the user to reveal it if not. For the Integrated Browser, open a fresh page withopenBrowserPage(forceNew: true); if the browser tools are missing, explain thatworkbench.browser.enableChatToolsmust be enabled and stop. - Dry-run the scenario without recording. Find stable role/name locators and check the assertions. Reset test data to a known start state.
- Write the complete script before recording, so no locators need to be discovered mid-recording.
Record in one call
Run the whole script in a single Playwright call so pauses between calls don't end up in the video. Allow enough timeout for the scenario.
Save the video and script in the current session's storage folder, not in
the workspace, and name them after the scenario, for example
<session folder>/add-and-complete-task.webm.
const viewport = await page.evaluate(() => ({ width: innerWidth, height: innerHeight }));
const even = v => Math.max(2, Math.floor(v / 2) * 2);
const scale = Math.min(1, 1280 / viewport.width);
await page.screencast.start({
path: '<session folder>/add-and-complete-task.webm',
size: { width: even(viewport.width * scale), height: even(viewport.height * scale) },
});
try {
await page.screencast.showActions();
await page.screencast.showChapter('Add a task');
const input = page.getByRole('textbox', { name: 'New task' });
await input.pressSequentially('Buy milk', { delay: 60 });
await input.press('Enter');
const task = page.getByRole('checkbox', { name: 'Buy milk', exact: true });
if (await task.isChecked()) {
throw new Error('New task was unexpectedly completed.');
}
await page.waitForTimeout(1000);
await page.screencast.showChapter('Complete the task');
await task.check();
if (!await task.isChecked()) {
throw new Error('Task did not become completed.');
}
await page.waitForTimeout(1500);
} finally {
await page.screencast.stop();
}
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 81 lines · 61 tokens per session scan A 0126d5892750
record-video is a skill published in the GitHub repository microsoft/vscode (194,353 stars, last pushed today), licensed MIT. It adds 61 tokens to every session and 818 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-29.
Other skills, from other repositories
supersplat
Use when editing 3D Gaussian Splatting (3DGS) .ply models in browser — cleanup noise/floaters, crop, transform, color-adjust, animate camera, publish online. SuperSplat: open-source browser-based 3DGS editor by PlayCanvas with PLY/compressed PLY/Splat/SOG/SPZ export.
posthog
Complete guide for the PostHog plugin — REST API access for querying analytics with HogQL, managing feature flags, inspecting events and persons, reading insights, experiments, cohorts, surveys, and more.
create-remotion-video
Create or edit video with Remotion. First-party clodex + Remotion skill. Contains full video-making process.
github
Complete guide for the GitHub plugin — REST API access for repositories, issues, pull requests, actions, releases, and search using a GitHub Personal Access Token.
supabase
Complete guide for the Supabase plugin — Management API access for running SQL queries, listing projects, managing edge functions, secrets, migrations, and inspecting project health.
javascript-sandbox
Best practices for using the clodex built-in JavaScript sandbox for browser debugging, fetched-data processing, attachments, and mini-app orchestration within its bundled capability boundary.