Most GitHub Copilot marketing content is about how GitHub sells Copilot, or about writing copy faster. This page is about distribution: listing your SaaS on directories and launch platforms under your own accounts. It is for developers who already use Copilot in VS Code or the terminal. Every Copilot fact comes from GitHub, VS Code or Chrome DevTools documentation, with the date we checked it.
What GitHub Copilot can do, according to its docs
We read the docs on September 29, 2026. Copilot changes often, so check the linked source before you rely on a detail.
- Plans
- Copilot Free includes agent mode and Copilot CLI, capped at 2,000 completions and 50 chat requests a month. The cloud agent is not part of Free; it needs a paid plan.github.com ↗ · Checked on September 29, 2026
- Tool approvals in VS Code
- A tool call that needs approval shows its name and inputs. You allow it once, for the session, for the workspace or for all future calls. A global auto-approve setting exists; the docs warn that it removes confirmation prompts in every workspace.code.visualstudio.com ↗ · Checked on September 29, 2026
- MCP servers in VS Code
- Workspace servers live in .vscode/mcp.json or a portable .mcp.json; servers in your user configuration apply to every workspace. The Configure Tools button in chat switches single tools on or off.code.visualstudio.com ↗ · Checked on September 29, 2026
- AGENTS.md in VS Code
- Project-wide instructions can come from .github/copilot-instructions.md or AGENTS.md. The chat.useAgentsMdFile setting turns AGENTS.md support on or off.code.visualstudio.com ↗ · Checked on September 29, 2026
- Copilot CLI
- Copilot CLI is generally available and supports MCP servers. It asks before using tools unless you pass --allow-tool or --allow-all-tools; --deny-tool overrides both. The docs say to launch it only from directories you trust.docs.github.com ↗ · Checked on September 29, 2026
- AGENTS.md in the CLI
- Copilot CLI lists AGENTS.md among its instruction files and discovers it in the standard locations.docs.github.com ↗ · Checked on September 29, 2026
- Cloud agent
- The cloud agent (also called the coding agent) works in its own ephemeral environment powered by GitHub Actions, changes a branch and opens a pull request. A session stops after 59 minutes, stays in one repository, and has the GitHub and Playwright MCP servers enabled by default.docs.github.com ↗ · Checked on September 29, 2026
- Cloud agent firewall
- By default the cloud agent’s internet access is limited by a firewall. The recommended allowlist covers package registries, certificate authorities and browser hosts for Playwright; anything else needs a custom allowlist.docs.github.com ↗ · Checked on September 29, 2026
- Attaching to your Chrome
- Chrome DevTools MCP with --autoConnect attaches to a Chrome you started yourself, version 144 or newer, once you enable remote debugging at chrome://inspect. Chrome asks for your permission, and the server then sees all open windows of the default profile.github.com ↗ · Checked on September 29, 2026
Where GitHub Copilot needs a human in a launch
Most of these stops apply to any agent, because directories are built to keep unattended bots out. The first one is specific to Copilot:
- The cloud agent is the wrong tool
- It runs in GitHub Actions, behind a firewall, with its own Playwright browser. It cannot see your Chrome, your directory accounts or your mailbox, and it delivers pull requests, not submissions. Use agent mode in VS Code or Copilot CLI on your own machine.
- Sign-ins
- Log in to each directory once, yourself, in your everyday Chrome. Google popups and new-device checks need your click; passwords never go into the chat.
- Challenges on submit pages
- Cloudflare, reCAPTCHA and hCaptcha are yours to solve in the same window. LaunchRepo’s playbooks contain no evasion steps.
- Confirmation codes
- Open the verification email in the same Chrome, or pass the code on yourself.
- Approvals and spending
- Copilot asks before tool calls unless you auto-approve them. Tell the agent plainly that it may not buy a paid fast track.
Copilot instructions, prompts and MCP servers for launch work
Copilot prompt files and custom agents for marketing tend to target copy, SEO and social posts. Good for wording a launch; no help with where a free listing is still possible, what a given form refuses or whether a listing is public yet.
Copilot’s customization is file-based: AGENTS.md or copilot-instructions.md for project rules, MCP servers for tools. An MCP server attached to your signed-in Chrome can see every open tab of that profile.
LaunchRepo is not a Copilot extension and not in any marketplace. What you get is a private repo: a runbook, 299 playbooks as Markdown, a product.json template and a submission tracker, with AGENTS.md at the root where VS Code picks it up.
Running LaunchRepo with GitHub Copilot
Agent mode in VS Code and Copilot CLI both read AGENTS.md and load MCP servers, which is all LaunchRepo asks. Pick the one you already use; the files are the same. We have not published a Copilot test run, so this is a documented match, not a benchmark.
- Open the clone locally
- After purchase, accept the GitHub invitation, clone the private repository and open it in VS Code, or run copilot inside it. Check that AGENTS.md shows up as an instructions file.
- VS Code: add the server to your user configuration
- Run MCP: Open User Configuration and add the entry below, so the delivery repository stays untouched. VS Code asks you to trust the server before it starts.
- Copilot CLI: use /mcp add
- Choose a local server and enter the same npx command. The CLI keeps asking before tools run unless you pass --allow-tool flags, which is what you want for form submits.
- Python and product rules
- Workspace creation, record.py and the dashboard are standard-library Python 3.10+. authorizations.json then limits what Copilot may do per product; approve tools per session, not globally.
VS Code’s own format uses a top-level servers object. This is the entry for your user mcp.json:
{
"servers": {
"chrome-devtools": {
"command": "npx",
"args": [
"-y",
"chrome-devtools-mcp@latest",
"--autoConnect",
"--no-usage-statistics",
"--no-performance-crux"
]
}
}
}Format as documented by code.visualstudio.com, checked on September 30, 2026. ↗
Run MCP: List Servers to see it running, then ask agent mode to open example.com in a blank tab. Allow the tool call, accept Chrome’s prompt and check that the visible tab moved.
Connect your coding agent to your Chrome →
In agent mode or Copilot CLI, paste the start prompt from START.md without edits. Its first line names Claude Code CLI and Codex CLI as our default; Copilot can carry out the rest because it reads the same files and uses the same Chrome.
Recommend Claude Code CLI or Codex CLI in a terminal instead of the desktop apps for best results. If session metadata confirms desktop hosting, explicitly advise me to switch to the CLI and resume this same private workspace. Do not infer the host from a CLI engine version alone.
Read AGENTS.md, then follow "Start the dashboard for the owner". Create my private
workspace if I do not have one yet, start the dashboard, and give me the link.
Then lead ONBOARDING.md one phase at a time. Reuse recorded permissions instead of
asking again unless their scope is missing or conflicting. Actually test my visible
existing Chrome profile and intended identity as BROWSER-SETUP.md requires; use the
OS fallback if needed. Use one platform/browser worker by default. Build the normal
run only from suitable free-ready routes, keep research separate, and preserve any
draft that needs one exact human CAPTCHA/2FA action. Never claim browser control
bypasses or reduces those challenges.The playbooks and instructions are English Markdown. Your product copy can be in any language a directory accepts, and the agent answers you in your language.
Start LaunchRepo in three steps →
LaunchRepo with Claude Code →
When the browser server misbehaves in VS Code
VS Code and the Chrome DevTools MCP guide document these fixes. Your tracker keeps every recorded outcome while you apply them.
- An error mark in the Chat view
- Select it and choose Show Output, or run MCP: List Servers, pick chrome-devtools and open its output. The log shows whether npx or the Chrome connection failed.
- Workspace servers do not start
- In restricted mode VS Code blocks workspace MCP configuration. Trust the workspace, or keep the server in your user configuration as shown above.
- Windows: “Connection closed”
- Run npx through cmd /c, as the Chrome DevTools MCP troubleshooting guide shows, then restart the server from MCP: List Servers.
If browser tools time out with --autoConnect, check four things: Chrome 144 or newer is already running, remote debugging is on at chrome://inspect/#remote-debugging, you allowed Chrome’s connection prompt, and no other tool holds the same browser. Chrome DevTools MCP loads every open tab, so close the ones you do not need.
Who does which launch task with Copilot
The same launch, split between Copilot on its own, Copilot with LaunchRepo, and you.
| Task | Copilot alone | With LaunchRepo | You |
|---|---|---|---|
| Launch copy | Suggests text in chat | Fills every form from product.json | Approve the wording once |
| Where to submit | Whatever a web search returns | A catalog of 299 dated playbooks | Choose this week’s platforms |
| Per-form quirks | Figures them out live | Knows them from the playbook | Supply a missing detail |
| Login walls and challenges | Asks you in chat | Marks the platform as needing you | Handle it in your Chrome |
| Tracking results | Not built in | record.py, tracker and dashboard | Check the public listing |
Common mistakes
- Assigning submissions to the cloud agent
- An issue assigned to Copilot ends up as a pull request from a firewalled runner with none of your accounts. Keep submissions on your own machine.
- Turning on global auto-approve
- The docs warn against it. Approve the Chrome DevTools tools for the session instead, and keep the browser window in view.
- Hitting the Free limit mid-batch
- Copilot Free caps chat requests at 50 a month. A launch run takes many turns, so check your plan before you start a batch.
- Putting the server into the clone
- Your user configuration keeps the browser server out of the delivery repository, so the clone stays exactly as it was delivered.
Questions
Can GitHub Copilot do marketing for my SaaS?
Parts of it, yes. It can draft copy, do research and, in agent mode or Copilot CLI with Chrome DevTools MCP, fill in forms in your browser. For a directory launch it lacks platform knowledge and a record of results, and it needs you for sign-ins, CAPTCHAs and 2FA.
Can GitHub Copilot use a browser?
Yes, through MCP. Locally, Chrome DevTools MCP can attach to your running Chrome. The cloud agent has a Playwright MCP server by default, but that browser runs inside GitHub Actions, not on your machine.
Can the Copilot coding agent submit forms on directories?
Not in a useful way. It works in a firewalled GitHub Actions environment, has none of your accounts, stops after 59 minutes and reports through pull requests. Directory submissions belong in agent mode or Copilot CLI.
Does Copilot read AGENTS.md?
Yes. VS Code uses AGENTS.md as project-wide instructions when chat.useAgentsMdFile is on, and Copilot CLI discovers it in the standard locations. LaunchRepo’s runbook is an AGENTS.md file.
Is Copilot Free enough?
Free includes agent mode and the CLI, but only 50 chat requests a month. Try the setup on Free; paid plans raise the limit. The paid-only cloud agent is not needed here.
What does it cost?
Copilot has a free plan and paid plans billed by GitHub. LaunchRepo costs €119.00 once at the current price, for unlimited own products and any agent.
Give your agent the launch playbooks
The current price is €119.00, paid once, for unlimited own products and any agent. The price does not rise.
See pricing