Product Hunt Launches MCP server

How an agent would call it

Two steps: search returns refs, then data_fetch opens one of them. The arguments below are the arguments that were sent.

{
  "tool": "search",
  "arguments": {
    "query": "developer-tools",
    "sources": [
      "producthunt"
    ],
    "limit": 2,
    "runId": "<from run_begin>"
  }
}

{
  "tool": "data_fetch",
  "arguments": {
    "sourceId": "producthunt",
    "ref": {
      "id": "1111197"
    },
    "runId": "<from run_begin>"
  }
}

This runs against an account — send Authorization: Bearer <token> with the call. Get one at app.bedrocktap.com/dashboard/tokens; the free tier needs no card, and it reaches producthunt like every other source.

What it costs. 0.45 credits per launch returned, and 0.45 per fetch. The search above returned 2 launches and cost 0.9 credits; the fetch cost 0.45.

What came back

{
  "ref": {
    "id": "1111197",
    "url": "https://www.producthunt.com/products/kilocode/launches/kilo-code-for-vs-code-3?utm_campaign=producthunt-api&utm_medium=api-v2&utm_source=Application%3A+PH+search+%28ID%3A+295453%29",
    "title": "Kilo Code for VS Code"
  },
  "text": "Kilo Code for VS Code\nParallel agents, diff reviewer, and multi-model comparisons\nWe've completely rebuilt Kilo Code for VS Code, built on OpenCode server. New portable core, parallel tool calls, subagent delegation, inline code review, multi-model comparisons. Get started: kilo.ai/install",
  "fetchedAt": "2026-09-12T14:58:02.149Z"
}

Each ref carried tagline, votes, comments, website, createdAt, topics, makers and description in its meta — the top launch in this topic had 896 votes, 239 comments, four topics and twelve named makers. The fetched payload returned the same launch record under the API’s own field names: name, tagline, description, votesCount, commentsCount, topics, makers, website and url.

Observed 2026-09-12. One call against a live source — ask again tomorrow and the rows move.

What a row contains

One row is one launch: its id, the launch page, and the product name. meta carries the one-line tagline, the vote and comment counts, the maker usernames, the product’s own website, the launch date, the topics it was filed under and the full description. The query is a topic slug, not free text — developer-tools, artificial-intelligence, productivity — because the API lists by topic and has no full-text post search, and a sentence would return nothing rather than the wrong thing.

A question it answers. Who else launched in this category recently, how do they describe it in one line, and how much attention did they get?

Routing guidance

What an agent reads to decide whether producthunt is the right source, and what to reach for when it is not.

Answers

Competitive landscape scans, what shipped recently in a category, finding makers in a niche, leaderboard rank over time, and review text with per-dimension ratings.

Query shape

A CATEGORY or DOMAIN for free-text discovery — "project management", "drones", "developer tools" — because free text resolves to topic slugs first. A specific product name usually resolves to nothing. A platform URL (leaderboard, topic, collection, product or maker) resolves that page directly. An absolute date window can be pinned instead of a rolling lookback.

Price

Priced in credits, per row, per query or per call depending on the source. Every call is quoted before it runs — `cost_estimate` returns the exact figure and spends nothing. See /pricing.

Legal basis

Public API — the platform’s own documented, openly available interface.

Latency

Fast (~3s) for a topic search; a date-window walk with no topic match can reach 40 seconds.

Prefer instead

`trustmrr` when the question is what a company earns rather than what it launched.

Cannot answer

Keyword search across post text — the platform’s search is client-side and not reachable. A specific product by name.

Caveat

The upstream has no full-text post search; queries resolve through a topic index, so category-shaped queries work and product-name queries do not.

Caveat

Complexity limits cap one walk, so deep back-to-back searches return partial coverage rather than errors. At the widest launch scope a long lookback yields only the newest day or two — narrow the scope when you want reach instead of volume.

Related sources