NicheDB

Sources in, feeds out: an open, ever-growing database of real-time public data, enriched.

CommunityOnlineNo sign-inGlobalFreeRead-only

What it can do

    What data it sees

    Do you need an account

    No: the server works without sign-in

    Sources in, feeds out: an open, ever-growing database of real-time public data, enriched.

    How to connect

    How to connect NicheDB to Claude

    The server address to paste into Claude settings:

    https://nichedb.dev/mcp

    1. Open Claude (claude.ai or the desktop app).
    2. Go to Settings → Connectors.
    3. Click Add custom connector.
    4. Paste the server address copied below into Remote MCP server URL and click Add.
    5. In a chat, click + → Connectors and switch the new connector on.

    Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

    Authorization

    After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

    How to connect NicheDB to ChatGPT

    The server address to paste into ChatGPT settings:

    https://nichedb.dev/mcp

    1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
    2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
    3. Open Settings → Connectors and click Create.
    4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
    5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
    6. In a new chat open + → Apps/Connectors and enable the connector.

    OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

    Authorization

    On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

    How to connect NicheDB to Cursor

    The server address to paste into Cursor settings:

    https://nichedb.dev/mcp

    Fastest: click Open in Cursor below and confirm the prompt.

    Manually:

    1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
    2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
    3. Save the file. The server appears in the MCP list; authorize it there if asked.

    Authorization

    If the service needs sign-in, an authorize button appears next to the server in the MCP list.

    Server tool list (21)

    Raw names from tools/list. Only developers need these.

    statsWhat this deployment holds: collections, sources, items, items added today, feeds.
    list_collectionsThe collections (niches) with their source, feed and item counts.
    list_adaptersEvery adapter this deployment can run, with the config fields each takes. Use before add_source.
    list_enrichersEvery enricher (YouTube videos, Wikipedia, repo stats, downloads, company profiles, TL;DRs) and which collections turn it on by default. Items carry their results under `enrichment`.
    list_sourcesSources with status, last run and item counts. Optionally one collection.
    list_feedsPublic feeds, most followed first. Optionally one collection.
    feed_itemsWhat a feed selects, newest first. Page with before_id.
    recent_itemsNewest items, optionally narrowed to a collection, source or kind.
    upcomingItems dated in the future (releases, launches, deadlines), soonest first.
    search_itemsFull-text search over titles, summaries and tags. Use this first when you know a name.
    match_itemsWhich title, channel or fixture is this name? Give a file name, a release name or a playlist entry ("Top.Gun.Maverick.2022.1080p.mkv", "US: ESPN2 HD"); it is cleaned (year, season and episode, quality tags, playlist decorations) and matched by similarity. A matchup ("NFL: Chiefs vs Bills", "Lakers @ Celtics", "Rangers at Celtic 19:45") is answered by the fixture whose two teams those are, kicking off between 36 hours ago and 7 days ahead unless date narrows it. Each result carries a score in 0..1. Use screen for films and TV, channels for television channels, sports for fixtures.
    get_itemOne item with its full adapter payload.
    create_feedSave a query as a feed (needs a key). Sources, kinds and tags narrow; q is a text match; upcoming keeps only future-dated items.
    follow_feedFollow a feed as the key owner (needs a key). Channels: webpush, email, webhook.
    submit_feedSuggest an RSS or Atom feed for the index. No key needed. An admin reviews every suggestion before anything is fetched; approved podcast feeds are handed to rssamplifier, everything else becomes a source here.
    search_profilesPeople: one entry per person assembled from every app that serves their OpenProfile.md (podcasters from p0dcasters, public profiles from OutreachGraph). Search by name, headline or anything in the document; newest change first. Each answer carries the page, the openprofile.md URL, accounts, topics, Broadcast shows and the Guest section.
    get_profileOne person by id, `<slug>-<id>` or handle: the parsed view and the OpenProfile.md as served.
    claim_profileClaim a profile as the key owner (needs a key). Proven when the key owner’s email is one the profile lists, or when the person’s site or show page links back to the profile’s nichedb page. An admin may claim on behalf of an email.
    update_profileEdit a profile you own (needs a key). Send `markdown`, a whole OpenProfile.md that replaces what you wrote before, or any of `name`, `headline`, `identity` (key: value, null removes), `sections` (name: body, `none` removes), `handle`, `public`. What you write wins over every source and survives every re-read.
    add_sourceAdd a source: an adapter with a config, fetched on a schedule (needs a key; admins and Pro). Call list_adapters first for the fields.
    run_sourceFetch a source now instead of waiting for its schedule (needs a key; owner or admin).