The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Dear Claude listing page.
MCP server that triggers local Claude Code instances from external platforms.
Say "Dear Claude" in Linear, GitHub, Jira, GitLab, Notion, or Obsidian — and a Claude Code instance spins up to handle it.
Your notes become architecture. Your tasks become pull requests.
Dear Claude is an MCP (Model Context Protocol) server that watches your project management tools for the phrase "Dear Claude". When detected, it spawns a local Claude Code instance that:
No Anthropic API keys needed. Works with your existing Claude Code subscription. 100% local and private — your code never leaves your machine.
| Platform | Trigger on issue/PR | Trigger on comment | Comment back | Emoji reactions | Sub-tasks | PR/MR review |
|---|---|---|---|---|---|---|
| GitHub | Yes | Yes | Yes | Yes | - | Yes |
| Linear | Yes | Yes | Yes | Yes | Yes | - |
| Jira | Yes | Yes | Yes | - | Yes | - |
| GitLab | Yes | Yes | Yes | Yes | - | Yes |
| Notion | Yes | Yes | Yes | - | - | - |
| Obsidian | Yes | - | Yes | - | - | - |
Instances from any platform get API access to all configured platforms. This enables workflows like:
That's it. Start Claude Code and Dear Claude is ready.
claude command available)bunx)If you prefer manual configuration, add to ~/.claude.json under mcpServers:
Then start Claude Code:
The MCP server starts automatically, sets up Tailscale Funnel, and prints your public webhook URLs.
Dear Claude uses Tailscale Funnel for stable public HTTPS URLs to receive webhooks. Setup is mostly automatic.
Install Tailscale:
Authenticate: tailscale up
Enable Funnel in the admin console: https://login.tailscale.com/admin/acls
"funnel" capability to your ACL policyThe server auto-configures Funnel on startup. Your public URL will be:
Tip: Run
tailscale serve status --jsonto verify your config. The health check auto-repairs the Funnel config every 10 seconds.
https://<your-hostname>.ts.net/dchttps://<your-hostname>.ts.net/dc/oauth/callback/githubhttps://<your-hostname>.ts.net/dc/webhook/githubhttps://<your-hostname>.ts.net/dc/setup/githubrepo, write:discussionGITHUB_ACCESS_TOKEN=ghp_...Note: With a PAT alone you won't get webhook-triggered instances. You'd use the
spawn_instanceMCP tool or/api/spawnendpoint instead.
| Environment Variable | Description |
|---|---|
GITHUB_CLIENT_ID | GitHub App client ID |
GITHUB_CLIENT_SECRET | GitHub App client secret |
GITHUB_WEBHOOK_SECRET | Webhook signature verification secret |
GITHUB_ACCESS_TOKEN | Personal access token (alternative to OAuth) |
https://<your-hostname>.ts.net/dc/oauth/callback/linearhttps://<your-hostname>.ts.net/dc/webhook/linearhttps://<your-hostname>.ts.net/dc/setup/linearAfter OAuth, only issues/comments from your authenticated Linear account trigger Claude.
Alternative: Use a Personal API Key (LINEAR_ACCESS_TOKEN=lin_api_...) from Linear Settings → Account → API.
| Environment Variable | Description |
|---|---|
LINEAR_CLIENT_ID | OAuth client ID |
LINEAR_CLIENT_SECRET | OAuth client secret |
LINEAR_WEBHOOK_SECRET | Webhook signing secret |
LINEAR_ACCESS_TOKEN | Personal API key (alternative to OAuth) |
https://<your-hostname>.ts.net/dc/webhook/jira
JIRA_WEBHOOK_SECRET, append it: ?secret=YOUR_SECRETissue_created, issue_updated, comment_createdClaude can create sub-tasks, transition issue status, and add comments via the Jira REST API v2.
| Environment Variable | Description |
|---|---|
JIRA_DOMAIN | Jira subdomain (e.g. mycompany for mycompany.atlassian.net) |
JIRA_USER_EMAIL | Your Atlassian email |
JIRA_API_TOKEN | API token from Atlassian |
JIRA_WEBHOOK_SECRET | Optional shared secret for webhook verification |
api, read_repository, write_repositoryhttps://<your-hostname>.ts.net/dc/webhook/gitlabGITLAB_WEBHOOK_SECRETFor self-hosted GitLab, also set GITLAB_URL=https://your-gitlab-instance.com.
| Environment Variable | Description |
|---|---|
GITLAB_ACCESS_TOKEN | Personal access token |
GITLAB_WEBHOOK_SECRET | Webhook secret token |
GITLAB_URL | GitLab instance URL (default: https://gitlab.com) |
NOTION_ACCESS_TOKEN=ntn_...https://<your-hostname>.ts.net/dc/oauth/callback/notionhttps://<your-hostname>.ts.net/dc/setup/notionNotion doesn't have native webhooks yet. To trigger Claude from Notion:
https://<your-hostname>.ts.net/dc/webhook/notionNOTION_WEBHOOK_SECRET if you want signature verification| Environment Variable | Description |
|---|---|
NOTION_ACCESS_TOKEN | Internal integration token |
NOTION_CLIENT_ID | OAuth client ID |
NOTION_CLIENT_SECRET | OAuth client secret |
NOTION_WEBHOOK_SECRET | Webhook verification secret |
Obsidian integration works via filesystem watching — no webhooks needed. Claude watches your vault for files containing "Dear Claude" and responds by appending to the same file.
.md file and save.Claude's response appears as a callout block appended to the same note. The frontmatter gets a claude-status field (processing → done / error).
How it works:
.md file changes in the vault.obsidian/, .trash/, and dotfile directories[[other-note]]) — Claude resolves and reads them| Environment Variable | Description |
|---|---|
OBSIDIAN_VAULT_PATH | Absolute path to your Obsidian vault |
OBSIDIAN_WATCH_DEBOUNCE_MS | Debounce delay in ms (default: 2000) |
Write "Dear Claude" (case-insensitive, with a space) anywhere in:
.md filesGitHub PR Comment:
Claude responds on GitHub:
Claude Instance Started (Instance:
abc12345) Processing your request...
Task Completed Found 2 issues:
- SQL injection in
user.ts:45- Missing input validation in
api.ts:102Created PR #15 with fixes.
Claude instances can spawn other instances for parallel work:
Claude will:
/api/spawn endpointWhen running as an MCP server inside Claude Code, these tools are available:
| Tool | Description |
|---|---|
list_platforms | List configured platforms and their status |
list_instances | List all Claude instances (filter by status) |
get_instance_status | Get detailed status of a specific instance |
get_instance_messages | Get conversation history for an instance |
kill_instance | Terminate a running instance |
get_running_instances | List currently running instance IDs |
spawn_instance | Spawn a new Claude instance for a task |
get_project_instances | List all instances in a project group |
The server also exposes REST endpoints on localhost:3334:
| Endpoint | Method | Description |
|---|---|---|
/health | GET | Health check + platform status |
/webhook/:platform | POST | Webhook receiver |
/api/instances | GET | List instances (?project_id= filter) |
/api/instances/:id | GET | Get instance details + children |
/api/instances/:id/kill | POST | Kill a running instance |
/api/spawn | POST | Spawn a new instance programmatically |
/api/platforms | GET | List configured platforms |
/setup/:platform | GET | Start OAuth flow |
/oauth/callback/:platform | GET | OAuth callback |
sudo systemctl start tailscaled && tailscale up (Linux)tailscale serve/tailscale funnel command may have overwritten it. The health check auto-repairs within 10 seconds. Verify with tailscale serve status --json.bun run src/index.ts status to verify the server is up. Test with curl https://<your-hostname>.ts.net/dc/health.issue_comment events. To trigger on a new PR, post a comment — PR descriptions alone won't trigger.https://<your-hostname>.ts.net/dc/setup/<platform> to re-authenticate.data/dear-claude.db and re-authenticate.claude) is installed and accessible in your PATH.data/workspaces/. Ensure write permissions.This repo ships a smithmark capability manifest at smithmark.yaml, alongside a static tool listing at tools.json. The manifest declares, in one place, the full external surface this MCP server touches when it runs:
api.github.com, github.com, api.linear.app, linear.app, gitlab.com, *.atlassian.net, api.notion.com, api.giphy.com, api.anthropic.com) and why.~/.dear-claude/**, data/**, ~/.claude.json, the debug log) and the access level.claude, tailscale, which, open, sudo, pkill).On every GitHub release, .github/workflows/smithmark-attest.yml produces a keyless Sigstore attestation over the published npm tarball: GitHub's OIDC token is exchanged for a short-lived Fulcio signing certificate, the attestation is signed with it, and the signature is recorded in the public Rekor transparency log. There are no signing keys and no secrets involved. The resulting smithmark-attestation.sigstore.json is attached to the GitHub release.
Verify a published version with:
(substitute the actual release tag for v1.1.0)
The manifest is the authoritative record of this server's capability surface. smithmark's lint is deliberately host-unaware and advisory: it flags every fetch() call site and every exec as "undeclared" regardless of what's in the manifest, so it will show findings on this codebase (and on any real MCP server) even though the egress above is fully declared. Treat lint output as a discovery aid, not a drift signal, until --strict or a host-aware successor ships.
MIT