Add Mcp From Catalog logo

Add Mcp From Catalog

Organization
speakeasy-api
add-mcp-from-catalog

Add a reviewed MCP Catalogue server to an explicit AICP project through the Speakeasy AI Control Plane Platform MCP.

Overview

Publisherspeakeasy-api
Repositorygram
Skill nameadd-mcp-from-catalog
Stars
269
Forks
33
Bundled files
Instructions only
LicenseAGPL-3.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by speakeasy-api on GitHub. Read the source before you install it.

Installation

Install the Add Mcp From Catalog AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/speakeasy-api/gram.git /tmp/gram
mkdir -p .claude/skills
cp -r /tmp/gram/server/internal/plugins/platform_mcp_skills/add-mcp-from-catalog .claude/skills/add-mcp-from-catalog
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Add Mcp From Catalog in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Add Mcp From Catalog on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Add Mcp From Catalog is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Add an MCP from the catalog

Use this workflow only through the authenticated Speakeasy AI Control Plane (AICP) Platform MCP. It follows the same guarded outcome a user would complete manually in the AICP dashboard: select a reviewed MCP, configure it for an explicit project, finish secure setup, and verify readiness. The package itself grants no organization access.

If the user instead supplies a remote MCP URL outside the reviewed catalogue, use the add-mcp-from-remote-url workflow.

Safety rules

  • Never ask the user to paste API keys, passwords, access or refresh tokens, OAuth codes, client secrets, secret headers, or MCP credentials into chat.
  • Never ask the user to supply an MCP endpoint. Use the server-owned catalog candidate and registration workflow.
  • Show non-secret provider and setup URLs only when a Platform MCP tool returns them.
  • Keep the chosen project and catalog candidate explicit. Do not infer either from a previous conversation or silently substitute another target.
  • Registration is private and does not distribute or publish the MCP.
  • Use send_platform_mcp_feedback only after asking for consent, and never include identifiers, URLs, credentials, payloads, headers, logs, or attachments.

Workflow

  1. Call list_projects to verify that the Platform MCP is authenticated and obtain the eligible projects. If authenticated discovery is unavailable, stop and ask the user to complete or repair AICP OAuth; do not claim that installation succeeded.
  2. Call search_mcp_catalog. Present the eligible projects and reviewed candidates, then ask the user to choose one exact project and one exact candidate.
  3. Call inspect_mcp_candidate for the selected candidate. Explain the bounded change and collect only the non-secret configuration fields declared by that result.
  4. After explicit confirmation, call register_catalog_mcp with the exact selected project, reviewed candidate, declared non-secret configuration, and a fresh idempotency key. Do not distribute it.
  5. Call get_mcp_readiness with the returned project and registration ID to inspect persisted readiness.
  6. Route secure setup from the exact readiness evidence:
    • For upstream_identity_provider_not_configured, explain that AICP can attach the one identity provider discovered from the persisted reviewed MCP source. Ask for explicit confirmation, then call attach_platform_mcp_identity_provider, present its exact Inspect authorization URL, and wait for the user to use Connect or Authorize.
    • For upstream_authorization_required, call attach_platform_mcp_identity_provider again with confirmation to retrieve the current server-issued Inspect authorization URL. Present that exact clickable URL and wait for the user to use Connect or Authorize.
    • For any other secure dashboard setup result, present only its exact server-returned setup URL. The user completes OAuth or secret entry outside the agent. Never request the resulting code, token, or secret in chat.
  7. After the user completes any secure handoff, branch by caller surface:
    • For a connected external Platform MCP client, call get_mcp_readiness with force: true. Do not rely on stale or inferred readiness.
    • For a managed project assistant, call get_mcp_readiness without force and report only the persisted actor-scoped evidence. A forced provider probe stays with connected external clients; identity-provider attachment does not.
  8. When readiness is current and ready, report the server-returned evidence. Registration is still private at this point: an MCP takes effect only once a plugin carries it.
  9. Call get_mcp_client_admission for the same project and registration ID and report the mode it returns together with the custom client ID metadata URLs it lists, which together decide which MCP clients may authorize against this server. Change it only when the user asks: explain what the proposed mode admits and refuses, ask for explicit confirmation, then call set_mcp_client_admission with that exact mode and confirmed: true. Known clients (presets) refuses an unlisted client at authorization with no fallback, so confirm the user accepts that before selecting it. Custom client ID metadata URLs are still added from the MCP server's Authentication settings in the AICP dashboard.
  10. Call list_plugins for the selected project, present the plugins it returns, and ask the user which one should carry this MCP. Do not choose for them and do not assume the default plugin.
  11. After the user names a plugin, call distribute_mcp_to_plugin with that exact plugin. A name matching nothing is refused as not_found and a name matching more than one as ambiguous_target; report the refusal and ask again rather than retrying with a different plugin, creating a plugin, or distributing to another project.

OAuth consent and approved secret entry are the expected out-of-agent stops. Project and catalog selection and identity-provider attachment confirmation stay in the conversation.

Frequently asked questions

What does the Add Mcp From Catalog AI skill do?

Add a reviewed MCP Catalogue server to an explicit AICP project through the Speakeasy AI Control Plane Platform MCP.

Why use Add Mcp From Catalog on TypingMind?

Because you install it once and use it with any model. Add Mcp From Catalog is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Add Mcp From Catalog in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/speakeasy-api/gram/tree/main/server/internal/plugins/platform_mcp_skills/add-mcp-from-catalog. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Add Mcp From Catalog?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Add Mcp From Catalog?

As many as you like. As long as a model supports skills, you can use Add Mcp From Catalog with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Add Mcp From Catalog AI skill free?

Yes. It is published on GitHub by speakeasy-api under the AGPL-3.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇