The social listening MCP server for Hacker News
Scored Hacker News mentions and live story and comment search, in the same feed as Reddit, X, LinkedIn, and Bluesky.
Summarize every comment on our Show HN post
Found 3 comments on your thread:
- High fit
hn_user_4821
"how does this compare to running your own scraper?"
- Medium fit
throwaway9911
"the pricing page could use more detail on limits"
- Medium fit
devtools_fan
"neat, bookmarking this for later"
Why this one
Every other Hacker News MCP server fetches stories. None of them monitor.
The open-source Hacker News MCP servers out there wrap the public HN API: get top stories, search stories, get a user's comment history. They're fine for what's on the front page right now, but they don't watch anything. Ask again tomorrow and you're back to square one.
SnitchFeed's Hacker News tools sit on top of a monitoring pipeline that scores every mention for relevance, the same way it already does for Reddit, X, and LinkedIn, and can put an HN mention next to a Reddit or X mention about the same thing in one answer.
| SnitchFeed | Open-source HN MCP servers | |
|---|---|---|
| Setup | Paste one link, OAuth | NPM or Docker install, manual config |
| Monitoring | Saved listeners, refreshed hourly | Stateless, fetches on request only |
| Filtering | AI relevance and sentiment scoring | None, raw stories and comments |
| Alerting | Slack, Discord, email, in addition to MCP | None, chat-only |
| Cross-platform | Reddit, X, LinkedIn, Bluesky, same connector | HN only |
Setup
Paste one link, OAuth
NPM or Docker install, manual config
Monitoring
Saved listeners, refreshed hourly
Stateless, fetches on request only
Filtering
AI relevance and sentiment scoring
None, raw stories and comments
Alerting
Slack, Discord, email, in addition to MCP
None, chat-only
Cross-platform
Reddit, X, LinkedIn, Bluesky, same connector
HN only
Built for Hacker News
Technical buyers, cross-referenced.
Hacker News threads are long, technical, and often the first place a product gets judged. Your agent reads them next to everything else.
One answer across every platform
HN mentions share a data model with Reddit, X, LinkedIn, and Bluesky, so your agent can compare the same launch or complaint across communities in one reply.
- Side-by-side comparisons in one question
- The same scoring on every platform
- One connector, one login
Search HN like a power user
Boolean syntax plus HN-specific qualifiers: author:/by: for a person, story: for one thread, points>/comments> for traction, all in the same query.
Pull your launch thread instantly
Search story:YOUR_STORY_ID and get every comment on your Show HN post back at once, no listener required.
Skip the threads nobody read
Add points>100 or comments>20 to any search, so your agent only surfaces threads that actually got traction.
- Filter by points or comment count
- Combine with author: or story: for precision
- Works as a one-off check, no listener needed
Try it
You ask for the outcome. Your agent handles the rest.
Copy any of these into Claude, ChatGPT, or Cursor once SnitchFeed is connected.
- “Show me HN threads where someone's asking for a tool like ours”
- “Did anyone mention us in an HN comment this week?”
- “Search HN comments for complaints about [problem]”
- “Summarize every comment on our Show HN post”
- “Group this thread's objections so we can answer them”
- “Which questions came up more than once?”
- “Show me HN and Reddit mentions of [competitor] side by side”
- “Is this week's HN complaint about us also showing up on X?”
- “Compare sentiment about [competitor] on HN and LinkedIn”
- “Start monitoring [competitor] on Hacker News”
- “Test this HN query before I save it”
- “Send HN mentions of our product to #launch”
Connected in under a minute
Paste one link into your AI app and log in with your SnitchFeed account. It stays connected from then on.
Server URL
https://api.snitchfeed.com/mcpPaste this into Claude under Settings, Connectors, or into ChatGPT, Cursor, or any app that supports MCP connectors.
Setup guide for each appClaude Code
claude mcp add SnitchFeed --transport http https://api.snitchfeed.com/mcpUsing Claude Code? One command connects it. Your browser opens a SnitchFeed login the first time, then it just works.
Skills
Ready-made playbooks for Hacker News
Skills chain these tools into finished work. Paste one into your assistant and it runs start to finish.
Reference
Everything your agent can do on Hacker News
What your agent can call behind the scenes. Useful if you're checking your client or writing your own prompts.
List your mentions
Pull scored HN mentions filtered by sentiment and date range, and cross-referenced with other platforms.
Search stories and comments
Search live Hacker News stories and comments by keyword, with author:, story:, and points> qualifiers.
Add a keyword
Add a new Hacker News keyword to a listener.
Tune or retire a keyword
Adjust or retire an existing Hacker News keyword.
Audit for noise
Flag noisy or wasteful Hacker News queries and suggest keeping, narrowing, slowing the scan, or retiring them.
Trend analytics
Trend HN mention volume and sentiment over time.
Common questions about Hacker News MCP
More questions about MCP in general? See the full FAQ at snitchfeed.com/features/mcp.
Still have questions? Reach out at hello@snitchfeed.com
Does this replace building on the HN API?
Can I combine Hacker News with other platforms in one question?
Can my agent search HN on demand?
How fresh is the HN data?
Do searches use my plan's credits?
Your next customer is judging you in a comment thread.
Connect SnitchFeed to the AI assistant you already use and ask it to find them.
More MCP guides
