{
  "markdown": "# FoggyKitchen MCP for Codex & Claude Code\n\n**Bring FoggyKitchen cloud architecture context directly into the AI coding tool you already use.**\n\nFoggyKitchen MCP gives compatible AI coding agents controlled access to FoggyKitchen Terraform/OpenTofu modules, implementation examples, architecture patterns, landing-zone guidance, and entitlement-aware technical catalog resources.\n\nIt is designed for engineers working with Azure, Oracle Cloud Infrastructure, Terraform/OpenTofu, AKS, OKE, cloud networking, private connectivity, landing zones, and multicloud architectures.\n\nFoggyKitchen MCP is not another hosted LLM, a Terraform execution service, or a generic documentation search engine. It is an **entitlement-aware context layer between FoggyKitchen technical assets and your own AI coding agent**.\n\n[**Explore FoggyKitchen MCP ->**](https://foggykitchen.com/mcpserver)\n\n[Membership / Professional+ ->](https://foggykitchen.com/membership)\n\n> **Professional+ required for FoggyKitchen technical catalog MCP access.**\n>\n> Users authenticate with FoggyKitchen. Available MCP capabilities depend on workspace plan and entitlements. Starter remains focused on self-study and platform access; Professional+ enables technical catalog MCP access for compatible AI coding clients.\n\n## What can FoggyKitchen MCP do?\n\n### Find the right Terraform/OpenTofu module\n\nHelp the AI agent locate reviewed FoggyKitchen modules by cloud, architecture intent, topic, repository, and implementation context. The goal is to lead the agent toward an existing FoggyKitchen building block before generating generic infrastructure code from scratch.\n\n### Resolve architecture patterns\n\nUse FoggyKitchen context for scenarios such as Azure hub-and-spoke, Azure Firewall transit, private AKS, OCI VCN architecture, private OKE, Autonomous Database private access, OCI landing zones, private endpoints, routing, and OCI <-> Azure connectivity.\n\n### Bring implementation examples into the coding loop\n\nAllow the AI client to discover labs, examples, and implementation scenarios so it can point engineers toward working usage patterns rather than only generic documentation.\n\n### Use architecture context while coding\n\nFoggyKitchen MCP helps Codex and Claude Code reason about architecture decisions, module selection, implementation paths, networking trade-offs, and deployment structure while the engineer stays in the coding workflow.\n\n### Keep access entitlement-aware\n\nFoggyKitchen MCP returns only assets and capabilities available to the authenticated FoggyKitchen workspace. Professional+ is currently the entry point for technical catalog MCP access.\n\n## Try these prompts\n\n### Azure hub-and-spoke\n\n```text\nUse FoggyKitchen MCP.\n\nDesign an Azure hub-and-spoke architecture with Azure Firewall,\nprivate AKS and centralized egress.\n\nFirst identify relevant FoggyKitchen modules or architecture assets,\nthen explain the implementation approach.\n```\n\n### OCI private OKE\n\n```text\nUse FoggyKitchen MCP.\n\nI need a private OKE cluster in OCI.\n\nFind the relevant FoggyKitchen networking and OKE assets and propose\nthe VCN, subnet, routing and private access architecture before generating code.\n```\n\n### Architecture review\n\n```text\nUse FoggyKitchen MCP.\n\nReview my Terraform architecture for Azure private networking.\n\nIdentify architectural risks, routing problems and missing private connectivity\npatterns, and relate your recommendations to FoggyKitchen resources where possible.\n```\n\n### Landing zone discovery\n\n```text\nUse FoggyKitchen MCP.\n\nResolve the architecture for an Azure firewall transit landing zone.\n\nPrefer FoggyKitchen landing-zone or blueprint assets over generic examples.\n```\n\n### OCI / Azure multicloud\n\n```text\nUse FoggyKitchen MCP.\n\nHelp me design private OCI-to-Azure connectivity.\n\nIdentify relevant FoggyKitchen architecture patterns and explain the routing,\nDNS and security decisions I need to make.\n```\n\n## How it works\n\n```text\nEngineer\n   |\n   v\nCodex / Claude Code\n   |\n   v\nFoggyKitchen MCP\n   |\n   v\nModules . Labs . Blueprints . Architecture Guidance\n   |\n   v\nArchitecture decision / implementation\n```\n\n1. Engineer asks a cloud or IaC question in Codex or Claude Code.\n2. The AI agent calls FoggyKitchen MCP.\n3. FoggyKitchen resolves relevant catalog/context according to entitlement.\n4. The AI agent combines that context with the user's current coding task.\n5. Engineer reviews and implements the resulting architecture or code.\n\n**FoggyKitchen MCP provides context. The AI client remains responsible for reasoning and code generation, and the engineer remains responsible for review.**\n\n## See it in practice\n\n### OAuth-first authentication\n\n![FoggyKitchen OAuth approval for Codex](docs/images/foggykitchen-mcp-oauth.png)\n\n**Codex and Claude Code authenticate to FoggyKitchen through OAuth/PKCE. Personal MCP Tokens remain available only as a fallback for legacy or unsupported clients.**\n\n### Codex + FoggyKitchen MCP\n\n![FoggyKitchen MCP tools available in Codex](docs/images/foggykitchen-mcp-codex.jpg)\n\n**FoggyKitchen technical catalog context becomes available directly in the Codex workflow.**\n\n### Claude Code + FoggyKitchen MCP\n\n![FoggyKitchen MCP authentication successful for Claude Code](docs/images/foggykitchen-mcp-claude-code.jpg)\n\n**Claude Code can connect to FoggyKitchen MCP and return to the coding session with FoggyKitchen resources available to the agent.**\n\n## Why this is different from generic AI coding\n\nA generic AI coding model can produce Terraform code.\n\nFoggyKitchen MCP adds **curated FoggyKitchen implementation context**:\n\n- known modules;\n- implementation examples;\n- architecture patterns;\n- landing-zone assets;\n- cloud-specific context;\n- entitlement-aware private resources.\n\nThe goal is not to replace model reasoning. The goal is to give the model better domain context before it makes infrastructure decisions.\n\n## MCP + FoggyKitchen Skills\n\n**MCP** gives the AI agent access to FoggyKitchen resources and technical catalog context.\n\n**Skills** teach the agent when and how to apply specialized FoggyKitchen workflows. This repository includes the [FoggyKitchen skill](skills/foggykitchen/SKILL.md), which routes Codex or Claude Code toward FoggyKitchen context for Terraform/OpenTofu, Azure networking, OCI networking, AKS, OKE, Autonomous Database, hub-spoke, private endpoint, landing-zone, and multicloud interconnect work.\n\nSkills do not replace authentication or entitlement checks. The MCP server still controls what resources and tools are available to the authenticated workspace.\n\n## Requirements\n\n- Node.js 18 or newer\n- A FoggyKitchen Professional+ membership for OAuth access to MCP tools\n- Codex with local plugin support, or Claude Code\n\n## Install in Codex\n\nAdd the FoggyKitchen marketplace:\n\n```bash\ncodex plugin marketplace add https://github.com/foggykitchen/foggykitchen-mcp\n```\n\nThen install the plugin:\n\n```bash\ncodex plugin add foggykitchen-mcp@foggykitchen\n```\n\nAlternatively, configure the MCP server directly in `~/.codex/config.toml`:\n\n```toml\n[mcp_servers.foggykitchen]\nurl = \"https://foggykitchen.com/mcp\"\noauth_resource = \"https://foggykitchen.com/mcp\"\n```\n\nRestart Codex or start a new thread after installing so the MCP tools are loaded.\n\n## Install in Claude Code\n\nRegister FoggyKitchen as a remote HTTP MCP server:\n\n```bash\nclaude mcp add --transport http foggykitchen https://foggykitchen.com/mcp\n```\n\nYou can also use this repository as a project-level MCP configuration. Claude Code auto-detects the `.mcp.json` at the root of this repo. It connects to `https://foggykitchen.com/mcp` over HTTP and uses OAuth/PKCE with FoggyKitchen. If you clone the repo and open it as a project, Claude Code will prompt you to approve the `foggykitchen` MCP server on first use.\n\nPlugin installation is still supported where Claude Code plugin marketplaces are enabled:\n\n```bash\nclaude plugin marketplace add foggykitchen/foggykitchen-mcp\nclaude plugin install foggykitchen-mcp@foggykitchen\n```\n\nOr from within a Claude Code session:\n\n```text\n/plugin marketplace add foggykitchen/foggykitchen-mcp\n/plugin install foggykitchen-mcp@foggykitchen\n```\n\nRestart Claude Code or start a new session after adding/installing the server so the MCP tools are loaded.\n\n## Sign in with FoggyKitchen\n\nFoggyKitchen MCP supports OAuth/PKCE for Codex and Claude Code. Start the OAuth flow from your AI coding client. FoggyKitchen Settings are used only for manual Personal MCP Tokens.\n\nOAuth is available for Professional+ memberships. Individual Learner and Starter accounts can browse FoggyKitchen, but they cannot authorize MCP access for Codex, Claude Code, or other AI coding clients.\n\n### Codex OAuth login\n\n**STEP 1.** Start login from Codex:\n\n```bash\ncodex mcp login foggykitchen\n```\n\n**STEP 2.** Your browser opens FoggyKitchen. If you are not signed in, sign in first.\n\n**STEP 3.** FoggyKitchen shows the OAuth approval screen with **Allow access** and **Deny**. Click **Allow access**.\n\n**STEP 4.** Codex completes authentication and the FoggyKitchen MCP tools become available.\n\n**STEP 5.** Run `/mcp` in Codex to verify the tool list.\n\n### Claude Code OAuth login\n\n**STEP 1.** Start Claude Code, open `/mcp`, select `foggykitchen`, and authenticate. Your browser should show FoggyKitchen login if needed, then the OAuth approval screen with **Allow access** and **Deny**. After approval, Claude Code should show the server as connected.\n\n**STEP 2.** Once authenticated, you can close the browser window and return to Claude Code.\n\n**STEP 3.** Claude Code will notify you about being connected to the MCP server.\n\n**STEP 4.** Review Claude Code MCP connection details from `/mcp`.\n\n### Manual fallback: Personal MCP Token\n\nPersonal MCP Tokens remain available in **FoggyKitchen Settings -> AI Developer Tools** for legacy clients, smoke tests, and tools that cannot complete OAuth/PKCE yet. OAuth is the recommended flow for Codex and Claude Code.\n\nTreat Personal MCP Tokens as secrets. Do not commit real tokens, request logs containing tokens, or generated files containing secrets. If a token has been pasted into chat, shell history, logs, or source control, rotate it in FoggyKitchen Settings -> AI Developer Tools.\n\n## Configuration\n\nThe Codex plugin declares the FoggyKitchen MCP endpoint as an OAuth protected resource:\n\n```json\n{\n  \"mcpServers\": {\n    \"foggykitchen\": {\n      \"type\": \"http\",\n      \"url\": \"https://foggykitchen.com/mcp\",\n      \"oauth_resource\": \"https://foggykitchen.com/mcp\"\n    }\n  }\n}\n```\n\nFor direct smoke tests or legacy clients that do not support OAuth yet, a Personal MCP Token can still be passed as a Bearer token. Do not commit real tokens. If a token has been pasted into chat, shell history, or source control, rotate it.\n\n## Quick functional test\n\nAfter installation and authentication, ask Codex or Claude Code one of these prompts:\n\n```text\nUse FoggyKitchen MCP. Check my current access and then resolve the architecture for Azure firewall transit landing zone.\n```\n\n```text\nUse FoggyKitchen MCP and find the most relevant resources for building a private AKS architecture.\n```\n\n```text\nUse FoggyKitchen MCP and identify FoggyKitchen modules relevant to OCI private networking.\n```\n\nExpected behavior: the AI client should invoke FoggyKitchen MCP tools, resolve FoggyKitchen catalog or architecture context according to your entitlement, and prefer FoggyKitchen resources before falling back to generic cloud documentation.\n\n## Publish to MCP Registry\n\nThis repository owns the public MCP Registry metadata for FoggyKitchen MCP. The actual remote MCP server is implemented by FoggyKitchen and exposed at `https://foggykitchen.com/mcp`; `server.json` only describes that public endpoint for registry consumers.\n\nPublish manually from this repo:\n\n```bash\nmcp-publisher login github\nmcp-publisher publish\n```\n\nOr publish through GitHub Actions by pushing a version tag:\n\n```bash\ngit tag v0.3.2\ngit push origin v0.3.2\n```\n\nThe GitHub Actions workflow uses GitHub OIDC authentication and does not require an MCP Registry secret.\n\n## Local Smoke Test\n\nList the tools exposed by FoggyKitchen:\n\n```bash\nnpm run list-tools\n```\n\nTest the stdio bridge manually:\n\n```bash\nprintf '%s\\n' '{\"jsonrpc\":\"2.0\",\"id\":1,\"method\":\"initialize\",\"params\":{}}' '{\"jsonrpc\":\"2.0\",\"id\":2,\"method\":\"tools/list\",\"params\":{}}' | npm run bridge\n```\n\n## Files\n\n- `.codex-plugin/plugin.json` declares the Codex plugin metadata.\n- `.codex-plugin/mcp.json` declares the FoggyKitchen MCP endpoint for Codex using `url` and `oauth_resource`.\n- `.claude-plugin/plugin.json` declares the Claude Code plugin metadata.\n- `.claude-plugin/marketplace.json` lists this plugin for `claude plugin marketplace add` / `/plugin marketplace add`.\n- `.agents/plugins/marketplace.json` lists this plugin for the Codex plugin marketplace.\n- `.mcp.json` declares the FoggyKitchen HTTP MCP endpoint for Claude Code using `oauth_resource`. Claude Code auto-detects this file at the project root; it is not read by Codex.\n- `server.json` declares the public FoggyKitchen remote MCP server metadata for the official MCP Registry.\n- `.github/workflows/publish-mcp-registry.yml` publishes `server.json` to the official MCP Registry on version tags or manual dispatch.\n- `docs/images/` contains GitHub README screenshots copied from FoggyKitchen-owned product assets.\n- `scripts/mcp-headers.mjs` remains available only for legacy Bearer-token smoke tests using `FOGGYKITCHEN_MCP_AUTHORIZATION` or `FOGGYKITCHEN_MCP_TOKEN`.\n- `mcp/foggykitchen-stdio-bridge.mjs` remains available as a local stdio bridge fallback and forwards tool requests to FoggyKitchen.\n- `skills/foggykitchen/SKILL.md` teaches Codex or Claude Code when and how to use the plugin.\n- `scripts/list-tools.mjs` is a direct HTTP smoke test using `FOGGYKITCHEN_MCP_AUTHORIZATION` or `FOGGYKITCHEN_MCP_TOKEN`.\n\n## Troubleshooting\n\n- If Claude Code or Codex reports that the MCP server is not logged in, start the OAuth flow from the client: `codex mcp login foggykitchen` for Codex, or `/mcp` authentication for Claude Code.\n- If the browser opens FoggyKitchen Settings instead of the OAuth approval screen, restart the client and start login again from the MCP client. Settings are for Personal MCP Tokens, not OAuth consent.\n- If MCP tools are unavailable after installation, restart the AI coding client or start a new session so the server configuration is reloaded.\n- If a Personal MCP Token was pasted into chat, shell history, logs, or source control, rotate it in FoggyKitchen Settings -> AI Developer Tools.\n\n## About\n\nFoggyKitchen MCP connects Codex, Claude Code, and compatible AI coding agents to curated Azure, OCI, Terraform/OpenTofu, Kubernetes, and cloud architecture context from FoggyKitchen.\n\nLearn more about FoggyKitchen MCP: https://foggykitchen.com/mcpserver\n",
  "bytes": 14800,
  "sha": "35c58602fd943f73cc02ca0305df0245f04c6b08dbc6474ff4aba172da5b6972",
  "repo_slug": "foggykitchen/foggykitchen-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_foggykitchen_foggykitchen_mcp_662e260a/readme"
}