{
  "markdown": "# Gildara MCP\n\n[![smithery badge](https://smithery.ai/badge/gildara-io/mcp-server)](https://smithery.ai/servers/gildara-io/mcp-server)\n\n**One prompt vault. Every agent. Zero setup.** Give Claude, Cursor, Windsurf, and Claude Code a shared, searchable prompt library that survives across chats, machines, and tools.\n\n```json\n{ \"mcpServers\": { \"gildara\": { \"command\": \"npx\", \"args\": [\"-y\", \"@gildara/mcp-server\"] } } }\n```\n\nPaste into `claude_desktop_config.json`, restart Claude — your Gildara tools appear in Claude's tool list. **No signup. No API key.** The server auto-provisions a free agent on first run and prints a one-click link code to pair it with your account later.\n\n![Gildara's tools surfaced in Claude](https://raw.githubusercontent.com/anaranillc/promptvault-ai-complete/master/docs/images/4-17-26-mcp-tools-in-claude.png)\n\n### → [Get started at gildara.io](https://gildara.io) — pair your auto-provisioned agent to sync your vault and unlock Pro.\n\n---\n\n## What your AI can do\n\n- **Search prompts semantically** — find by meaning, not just keywords (Gemini embeddings).\n- **Resolve + run prompts** — compiled system prompts with operating-contract sections assembled, auto-repair on JSON schema failures.\n- **Create + append** — save new prompts, or append persistent facts to memory-type prompts.\n- **Dispatch briefs to agents** — say *\"send this to gevorg\"* and the MCP server creates a Brief, ships it as a `.md` attachment to your paired Telegram chat, and @mentions the target agent's bot. One call closes the sketch-in-chat → agent-on-another-box → result-in-chat loop. See [docs/briefs/dogfood-rollout-v0.md](https://github.com/anaranillc/promptvault-ai-complete/blob/master/docs/briefs/dogfood-rollout-v0.md) for the agent-side integration recipe.\n- **Browse 48+ blueprints** — pre-built agent templates.\n\n## Setup\n\n> **No API key required on first run.** As of v0.6.0, the MCP server auto-provisions a new agent account on first startup if no `GILDARA_API_KEY` is set, and caches the key at `~/.gildara/auto-key.json` for future runs. Skip to \"Add to Claude Desktop\" below — zero curl required.\n>\n> On first run, the server prints a **link code** to stderr (visible in Claude Desktop's MCP logs). Visit `https://gildara.io/link?code=XXXXXX` to pair the auto-provisioned agent with your human account and inherit your subscription tier.\n\n### 1. (Optional) Use an existing API key instead of auto-provisioning\n\nSkip this whole section if you're happy with auto-provisioning.\n\nOtherwise, get a key one of these ways:\n\n**Manual provision via API:**\n\n```bash\ncurl -X POST https://gildara.io/api/v1/provision \\\n  -H 'Content-Type: application/json' \\\n  -d '{\"agent_label\": \"my-agent\"}'\n```\n\n**From the web:** Go to [gildara.io/account](https://gildara.io/account) and create an API key with the **agent-standard** preset.\n\nThen set `GILDARA_API_KEY` in your MCP config (see examples below).\n\n### 2. Add to Claude Desktop\n\nEdit your `claude_desktop_config.json`. **No env block needed for auto-provision:**\n\n```json\n{\n  \"mcpServers\": {\n    \"gildara\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@gildara/mcp-server\"]\n    }\n  }\n}\n```\n\nOr, with an existing key:\n\n```json\n{\n  \"mcpServers\": {\n    \"gildara\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@gildara/mcp-server\"],\n      \"env\": {\n        \"GILDARA_API_KEY\": \"pvk_your_key_here\"\n      }\n    }\n  }\n}\n```\n\n### 3. Add to Cursor\n\nOne click:\n\n[![Install MCP Server](https://cursor.com/deeplink/mcp-install-dark.svg)](https://cursor.com/install-mcp?name=gildara&config=eyJjb21tYW5kIjoibnB4IiwiYXJncyI6WyIteSIsIkBnaWxkYXJhL21jcC1zZXJ2ZXIiXX0%3D)\n\nOr manually: edit `.cursor/mcp.json` in your project root using the same configuration as Claude Desktop (with or without the `env` block — same rules apply).\n\n### 4. Add to Claude Code\n\nRun this in your terminal:\n\n```bash\nclaude mcp add gildara -- npx -y @gildara/mcp-server\n```\n\nAuto-provision works here too. Set `GILDARA_API_KEY` in your environment only if you want to use an existing key.\n\n### 5. Add to Claude Chat (claude.ai)\n\nGildara supports **Remote MCP** via HTTP/SSE.\n\n1. Go to **Settings > Connectors** in Claude Chat.\n2. Select **\"Add custom connector\"**.\n3. **Name:** Gildara Vault.\n4. **Remote MCP server URL:** `https://gildara.io/api/mcp/mcp`.\n5. **Authentication:** OAuth 2.1 (you will be redirected to the Gildara consent page to authorize).\n\n## Tools\n\n| Tool | Description |\n| :--- | :--- |\n| `search_prompts` | Semantic search over your vault using Gemini embeddings |\n| `list_prompts` | List all prompts in your vault |\n| `get_prompt` | Get prompt details and operating contract config |\n| `resolve_prompt` | Get compiled system prompt with variable substitution |\n| `run_prompt` | Run prompt through AI with auto-repair |\n| `create_prompt` | Create a new prompt |\n| `append_memory` | Add info to an existing memory-type prompt |\n| `save_brief` | Dispatch a brief to a named agent (MD attachment via Telegram, @mention routes to the right bot) |\n| `list_blueprints` | Browse 48+ agent blueprint templates |\n| `get_account_link` | Get the URL to link this agent's vault to a human Gildara account |\n\n## Example usage\n\nOnce connected, just tell your AI:\n\n- \"Search my vault for prompts related to security audits.\"\n- \"Resolve my code-review-agent prompt with the `code_diff` variable.\"\n- \"Run the security-audit prompt against this code.\"\n- \"Append this preference to my 'Personal Context' memory: I prefer TypeScript over JavaScript.\"\n- \"Send this to gevorg: analyze last week's Meta Ads performance and flag the bottom 3 campaigns.\"\n\n## Local-First Support\n\nGildara features a local-first architecture. The MCP server automatically caches resolved prompts to `~/.gildara/cache/` to ensure your agents remain functional even if the Gildara API is temporarily unreachable.\n\n## Environment variables\n\n| Variable | Required | Description |\n| :--- | :--- | :--- |\n| `GILDARA_API_KEY` | No | Your API key (`pvk_...`). If unset, the server auto-provisions a new agent account on first run and caches the key at `~/.gildara/auto-key.json`. Set this explicitly to use an existing key. |\n| `GILDARA_BASE_URL` | No | Override API URL (default: `https://gildara.io`). |\n\n## License\n\nMIT\n\n---\n\n## Contributing / Maintainers\n\nThis package is the **stdio** MCP server. A sibling **HTTP** server lives at `app/api/mcp/[transport]/route.ts` in the parent monorepo (powers the Claude.ai native connector). Both must stay at parity — tool names, argument schemas, response text.\n\nBefore adding, removing, or changing a tool here, read [`docs/MCP_INTEGRATION_CONTRACT.md`](../docs/MCP_INTEGRATION_CONTRACT.md) and run `npm run test:unit` from the monorepo root to verify parity.\n",
  "bytes": 6702,
  "sha": "cbafc1b766fbed104309986e148936c4d13f36eff6b5130aa59c97b07579ba1e",
  "repo_slug": "gildara-io/mcp-server",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_gildara_mcp_server_5e8f9e48/readme"
}