MCP Server

Answers From Me runs a remote Model Context Protocol (MCP) server, so AI tools such as Claude, ChatGPT, Claude Code and Google Gemini can work with Answers From Me as a set of callable tools, without writing any integration code. It serves two audiences:

  • Anyone can consult experts: ask an expert's knowledge agent a question from inside the conversation they're already having.
  • Experts can also run their own agent: catch up on activity, fill knowledge gaps, add content, revise answers and invite seekers. See Manage Your Agent from Claude or ChatGPT for what that looks like day to day.

What is Model Context Protocol (MCP)?

MCP (Model Context Protocol) is an open-source standard for connecting AI applications to external systems. Using MCP, AI applications like Claude or ChatGPT can connect to data sources, tools, and workflows--enabling them to access key information and perform tasks.

Why access an agent from within an agent?

General-purpose AI tools powered by LLMs are good at recalling, synthesizing, and summarizing information. But as answer engines they are often flawed, especially on topics where taste or opinion is important: they equivocate, present the median opinion of over-weighted sources such as Reddit, have invisible biases, and when pressed for specifics they may hallucinate or make up citations.

An expert's agent answers differently. It draws on one real person's knowledge, experience, and point of view, so it can give you the opinionated, specific take a generalist won't. Pulling that agent into your conversation--for example, @@expert_handle What's your perspective on this?--drops the expert's voice into the work right where you need it, alongside everything the assistant already has in context.

It also closes the loop for the expert. When their agent is consulted, the expert sees the questions being asked of their knowledge and how it is being applied, giving them a window into where their expertise is useful and a chance to sharpen the agent's answers over time. Experts can do that sharpening from the same AI tool, using the expert tools below.

In short, the Answers From Me MCP server allows you to add human intelligence to your AI workflows.

MCP connection basics

Any MCP-capable tool can connect to Answers From Me with a single URL and a one-time sign-in using OAuth (or an API key).

The server is hosted at:

https://api.answersfrom.me/mcp

It speaks Streamable HTTP, the standard transport supported by remote-MCP clients.

Signing in with OAuth works on every plan, including Free, and for seekers as well as experts. API keys are part of the Basic plan and higher. Your plan's usage limits apply to everything you do through the server, as they do in the web app (see Plans, Usage and Limits).

Tools

Which tools your AI tool sees depends on the account you sign in with. Please note that this server is in beta, and tools and descriptions may change as we receive user feedback.

For everyone (8 tools before you become an expert)

  • ask_expert — Send a question to an expert's knowledge agent and get an answer grounded in that expert's own knowledge. Each answer ends with a response_id that the host passes back to continue the same conversation, so follow-up questions keep their thread without repeating earlier context.
  • get_answer — Fetch the current version of an earlier answer: one the expert was still reviewing, or one the expert has since revised.
  • list_experts — List the experts you can already reach: your own agent plus any agents you subscribe to.
  • search_experts — Search the public Answers From Me expert directory by name, or by topic or need.
  • add_expert and remove_expert — Subscribe to an expert, or unsubscribe.
  • afm_help — Answer a question about Answers From Me itself, from this help center.
  • become_expert — Start your own expert account and agent, usually after the Become an expert interview (below). Shown only to accounts that aren't experts yet.

For experts (22 tools)

Experts get everything above except become_expert, plus tools for running their own agent:

  • Content
    • add_content — Add text, a link, or an uploaded file to your knowledge base under a content key. Adding under the same key replaces the old version.
    • create_upload — Upload a file the AI tool can read itself, such as a file on your computer in Claude Code.
    • upload_from_device — Show a file picker in the chat so you can upload a file from your device. The picker appears only in hosts that support MCP Apps; in other hosts, the AI tool uses create_upload for a file it can read, or you upload on the Knowledge page.
    • list_content and get_content — List your content with its status and key, and read back a text document.
    • remove_content — Remove content added through MCP or the REST API, by its key. Asks you to confirm first.
  • Activity and usage
    • get_activity — Catch up: conversations, answers and new seekers since your last check, plus knowledge gaps, flagged and held answers, failed sources and limits near their cap.
    • get_usage — Your plan and how much of each limit you've used.
  • Conversations and Q&A
    • list_conversations and get_conversation — Read seekers' conversations with your agent.
    • revise_answer — Send your own answer to a seeker in place of your agent's, or approve a held answer with your text. Asks you to confirm first.
    • save_qa and list_qa — Save a question and your answer as a saved Q&A, and list them.
  • Seekers
    • list_seekers — List your seekers with their status and last activity.
    • invite_seeker — Email someone an invite to join your audience. Asks you to confirm first.

Content you added in the web app has no content key, so the expert tools can read it but not change or remove it. Some management, such as agent settings and deleting web-app sources, is in the Expert REST API only, and some stays in the web app. Manage Your Agent from Claude or ChatGPT has the full list.

Guided workflows

The server also offers guided workflows as MCP prompts. AI tools that support prompts list them alongside the connector, and Claude Code shows them as slash commands. Where your tool doesn't list them (ChatGPT, for example), ask for the same thing in plain language; the tool descriptions carry each workflow.

Workflow Who sees it What it does
Ask an expert (ask) Everyone Asks an expert a question
Review a draft with an expert (review_draft) Everyone Has an expert review the draft in your conversation
Become an expert (interview_me) Non-experts Interviews you, then sets up your expert account and agent
Weekly review (weekly_review) Experts Catches up on the week, then fixes gaps and flagged answers one at a time
Fill knowledge gaps (fill_gaps) Experts Drafts content for unanswered questions from your own material
Write content together (co_write) Experts Writes or revises a document in your knowledge base with you
Check usage (usage_check) Experts Shows which limits are near their cap and your options

Use @@ to Ask an Expert

Our MCP server gives the LLM guidance to direct prompts that start with @@expert_handle to the expert's agent, while @@ (question) will first list the available experts and then ask the most relevant one. Remember @@ as meaning ask the expert's agent.

Adding a Custom Connector

This section outlines the basics of the one-time connection process for adding the Answers From Me MCP server to your AI tool. The exact steps vary by tool, so check your tool's documentation for specific instructions on adding a remote MCP server.

Web and desktop clients such as Claude and ChatGPT can connect to the Answers From Me server as a custom connector, authenticating through OAuth instead of a secret key. Sign in with your Answers From Me account in a browser when prompted, and the client manages the token from there, with no key to copy or store. This is the preferred approach for these clients.

In the client, add a custom connector and provide a name and the server URL:

https://api.answersfrom.me/mcp

Leave the OAuth Client ID and Client Secret fields empty. The Answers From Me server registers the client automatically.

After adding the connector, the client walks you through signing in to Answers From Me and authorizing access. Once authorized, the Answers From Me tools are available in the client. If you're an expert, sign in with the account that has your expert agent to get the expert tools.

OAuth Alternative - API Bearer Token

OAuth is preferred for the Claude and ChatGPT web and desktop clients, and it works on every plan. Command-line tools that support OAuth sign-in for remote MCP servers can use it the same way: add the server URL without a header and sign in when the tool prompts you. For tools that don't, use an API key as described below.

The MCP server uses the same secret keys as the API. To create one, sign in to the app, click Settings in the left navigation of your Expert Dashboard, click the Connect tab, and scroll to the API Keys card. Turn on Enable API keys, then select the Create new secret key button. API keys are part of the Basic plan and higher (see Plans, Usage and Limits). The secret key will only be presented once, so make sure to copy it and put it in a safe place. You can create as many keys as you want for different applications, and delete a key if it is no longer used.

The same secret key works for the MCP server, the Open Responses API and the Expert REST API. A sign-in made through OAuth works with the MCP server only.

Connecting from Claude Code

Authentication uses your secret key as a Bearer token in the Authorization header. To register the Answers From Me server with Claude Code, run:

claude mcp add --transport http afm 'https://api.answersfrom.me/mcp' \
    --header "Authorization: Bearer $AFM_API_KEY"

(Where $AFM_API_KEY is the secret key you created in the previous step and is loaded into your environment.)

After adding it, Claude Code can call the Answers From Me tools like any other tool, and the guided workflows appear as slash commands. Other MCP hosts follow the same pattern: a remote (HTTP) MCP server at https://api.answersfrom.me/mcp with the header Authorization: Bearer (your secret key).

Examples

You don't call these tools directly. You ask for what you want in plain language, and the tool you're working in decides which tool to call. A few prompts that work well:

"Find me an AFM expert on service design" — searches the public Expert directory and comes back with matching Experts, their names, and their handles.

"Consult my most relevant AFM expert about this pricing page" — lists the Experts you can already reach, picks the best fit, and asks them.

"@@scott, can you review my resume (attached)?" — goes straight to Scott's Agent. Whatever you're working on in the conversation goes with the question, so the Expert answers on your actual material rather than a description of it.

"Ask @@scott a follow-up: what would you change first?" — stays on the same thread with Scott's Agent, so you don't have to restate the context.

"@@ what am I missing here?" — a bare @@ lets the host pick from the Experts you can reach.

"What's happened with my AFM agent this week?" — for experts, catches up on conversations, gaps and flagged answers.

Naming an Expert with @@ is the most direct way to get the answer you want. Without it, the host decides whether an Expert is worth consulting and which one to ask, so being specific about the topic — and that you want an Expert — makes it more likely to reach for one.

Advanced Configuration — Advising the LLM on When to Consult Experts

By default, the server advises the host to consult an Expert only when you explicitly ask for one. The optional invocation_mode query parameter changes the advice the server gives, which you set once when you register the server:

  • explicit (default) — Ask an Expert only when you request one with @@ or by name.
  • suggest — Suggest consulting a relevant Expert, but wait for you to agree before asking.
  • proactive — Consult an Expert without being asked when an Expert's answer would materially improve the response, and say so when it does.

Set the mode by adding it to the registered URL. For example, to let the host suggest Experts:

claude mcp add --transport http afm 'https://api.answersfrom.me/mcp?invocation_mode=suggest' \
    --header "Authorization: Bearer $AFM_API_KEY"

This is guidance to the LLM rather than a rule it must follow — the host ultimately decides which tools to call and when. Treat the mode as a way to set expectations, not a guarantee.

Troubleshooting

  • Tools are missing, or haven't changed after an update. When the server is updated, a desktop app can keep the tool list from its earlier connection, including whether the file picker appears, until it's restarted. Restart the app and start a new conversation.
  • The expert tools don't appear. Check that you signed in with the account that has your expert agent. Disconnect and reconnect the connector to sign in again.
  • Anything else. Check your tool's documentation for custom connectors, or ask the afm_help tool, which answers from this help center.