Connect PostHog to your phone agent. Answer product and usage questions on the call.

PostHog is where your product data lives: events, feature flags, experiments and errors. With the PostHog MCP integration, your AgentCalls agent can look up that data during a call and answer plainly. A customer asks whether a feature is turned on for their account. A partner asks how many people used a flow this week. Your agent checks PostHog and gives a short, accurate answer, so your team does not have to stop what they are doing. The agent reads your PostHog data only. It does not change flags or experiments.

What your agent can do with PostHog MCP

  • Look up whether a feature flag is on or off for a given user or account
  • Report recent event counts and trends for a product flow
  • Check the status of a running or finished experiment
  • Tell a caller whether errors have been logged for a feature recently

PostHog MCP exposes 1 actions. You choose which ones the agent may use and whether each runs during or after the call.

Example calls

Customer asks if a feature is enabled

  1. Caller

    Hi, I can't see the new reports tab in my account. Is it turned on for us?

  2. Agent

    Let me check that for you. What email address is on the account?

  3. Caller

    It's maria@brightstudio.co.

  4. Agent

    Thanks. The reports feature flag is off for that account right now. Our team enables it in rollout groups, so I've noted your request and someone will confirm by Thursday.

Partner asks about weekly usage

  1. Caller

    We launched the onboarding checklist on Monday. How many people finished it so far?

  2. Agent

    Checking now. Since Monday, 342 users started the onboarding checklist and 218 completed it, which is about 64 percent.

  3. Caller

    That's better than I expected. Can you send that to me by email?

  4. Agent

    Yes, I'll send the summary to your address on file today.

Support lead checks for recent errors

  1. Caller

    Customers are saying exports are failing. Are we seeing errors on that?

  2. Agent

    Looking at error tracking for exports. There were 27 export errors in the last two hours, up from a usual 3 or 4.

  3. Caller

    Okay, I'll escalate that to engineering now. Thanks.

  4. Agent

    I've logged the spike and added a note for your on-call team.

How to connect PostHog MCP

  1. 1

    Create your agent

    Sign up to AgentCalls, pick an English or Arabic voice and write what the agent should know and do in plain language.

  2. 2

    Connect PostHog MCP

    From Integrations in the dashboard, choose PostHog MCP and enter the connection details from your PostHog MCP account settings.

  3. 3

    Choose the actions

    In the agent editor, enable the PostHog MCP actions the agent may use and set each to run during the call or afterwards. Then test it from your browser.

Questions, answered.

Does the agent work during the call or after it?

During the call. Your agent looks up PostHog data while the caller is on the line and gives a spoken answer. It can also send a summary afterwards if you set that up.

What do I need to connect PostHog?

A PostHog account and the credentials from your PostHog settings. Sign in to the AgentCalls dashboard, open Tools, choose PostHog and enter your credentials. Integrations are included from the Growth plan.

Can the agent change feature flags or experiments?

No. The PostHog integration reads your product data, flags, experiments and error tracking. It does not change them.

Page drafted from the tool's public description and reviewed automatically for accuracy of claims

Put your agent to work with PostHog MCP.

Connect in minutes, test from your browser, go live on a number in the US, UK or Gulf. Free to start, no card needed.