Connect Gitea to Gemini CLI via MCP: Issues and Files

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Gitea, paste your credentials, mint an MCP API key — done. No Docker, no git clone, no local server.

Gitea + Gemini

Let your AI read your Gitea: find repositories, triage issues and pull requests with their comments, files, diffs and reviews, check releases and organizations, and read branches, commits and files. gitea.com or your own Gitea server, with an access token. Read-only.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

GITEA_URL, GITEA_TOKEN

Step 1 — Get credentials

Setting up takes about 2 minutes. You need an account on gitea.com or on your own Gitea server, and an access token with Read permission for Repository, Issue, Organization and User.

1. In Gitea, click your avatar, then Settings, then Applications. Under Manage Access Tokens, type a name (for example "AnythingMCP") and set Repository, Issue, Organization and User to Read (the scopes read:repository, read:issue, read:organization and read:user). On a Gitea version without token scopes, the token gets the full access of its user.
2. Click Generate Token and copy the token right away: Gitea shows it only once. Paste only the token into GITEA_TOKEN, without the word token or Bearer in front. The connector adds token itself.
3. Set GITEA_URL to the address you open Gitea at, for example https://gitea.com or https://git.example.com, without /api/v1 and without a trailing slash. If Gitea runs under a sub path, keep it (https://example.com/gitea).
4. Install the connector and run gitea_get_current_user: it shows the user the token belongs to and confirms that address and token work.

Read-only: all 30 tools only read. The connector cannot create, change or delete anything in Gitea, and the AI sees exactly what the token's user can see.

What it reads: repositories (searched across the whole instance, or listed for your user, an organization or another user), issues with comments, labels and milestones, pull requests with their changed files, diff, commits and reviews, releases with notes and attached files, organizations with members and teams, branches, commits, the file tree and file contents. gitea_search_issues finds issues and pull requests across all repositories that are assigned to you, created by you or wait for your review.

Repositories and numbers: every repository takes owner (a user or organization) and repo, the two parts of https://host/owner/repo. Issues and pull requests share one number sequence per repository. Gitea treats a pull request as an issue too, so gitea_get_issue and gitea_list_issue_comments also work with a pull request number for its title, labels and conversation, while gitea_get_pull_request, gitea_get_pull_request_files and gitea_get_pull_request_diff give the code side.

Your own server: a Gitea on an internal host (a private IP or an internal name) needs a self-hosted AnythingMCP on the same network, with the hostname in SSRF_ALLOWED_HOSTS, or the outbound guard blocks the call. Forgejo, and Codeberg which runs it, offer the same API v1 routes, so an address such as https://codeberg.org should work too. The connector is written against the Gitea API.

Limits: a Gitea server returns 30 items per page by default and at most 50 (its MAX_RESPONSE_ITEMS setting), so the AI asks for the next page instead of a bigger limit. Pull request diffs and file contents over 200 KB are cut: open the pull request's html_url for the rest.

Common errors: 401 means the token is wrong, expired or deleted, or GITEA_URL points at another server. 403 means the token lacks a scope the call needs (Gitea names it in the message), for example read:issue for issues or read:organization for organizations. 404 means the repository, issue or file does not exist or the user cannot see it: Gitea answers 404 for private resources it hides. 422 names a missing or invalid field.

The connector does not create or edit issues, comments, pull requests, releases, branches or files, does not merge or approve, and does not cover Actions runs and workflows, packages, wikis, notifications, webhooks, deploy keys or administration.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Gemini

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Gemini CLI reads MCP servers from ~/.gemini/settings.json (on Windows %USERPROFILE%\.gemini\settings.json). Add:

{
  "mcpServers": {
    "anythingmcp": {
      "httpUrl": "https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID",
      "headers": { "X-API-Key": "YOUR_MCP_API_KEY" }
    }
  }
}

1. Create an MCP API key in AnythingMCP under MCP Servers → your server → API keys, and paste it in place of YOUR_MCP_API_KEY.
2. Save the file and restart gemini.
3. Run /mcp inside Gemini CLI (or gemini mcp list in your shell) — anythingmcp should show as connected.
4. Start prompting — Gemini CLI can now call the Gitea tools.

Available tools

FAQ

Where does Gemini CLI look for MCP servers? In ~/.gemini/settings.json for your user, or in .gemini/settings.json inside a project. gemini mcp list shows each server and whether it connects.

Next steps

• Connector landing page
• Connect to Claude
• Connect to ChatGPT