buzzabout__collect_mentions
Collects public posts for a topic.
LABSCO SUMMARY
Nothing here deletes, but the collection tools spend credits.
The tools collect, cluster, query and save posts into your collections; none removes anything. What a careless agent spends is credits: collecting mentions or audiences, saving a post and detecting patterns all charge for data, and asking the assistant is billed for model use. Let the free lookups run freely and keep the collection tools behind your approval.
What it is
Buzzabout's hosted server over its social-listening data, with three kinds of tool: research tools you drive to collect, cluster and query mentions; an assistant you hand a whole job to; and lookups over what the account already has.
What you get
- buzzabout__collect_mentions, buzzabout__collect_audience and buzzabout__detect_patterns — collect the conversation, profile its authors and cluster it; buzzabout__query asks your own questions of the result
- buzzabout__ingest_mention — save one public post into a collection by its URL
- buzzabout__ask — a prompt against chosen datasets, optionally referencing specific posts, capped at 50
- buzzabout__get_message — retrieve the answer for a chat and message id
- buzzabout__render — render a block reference
- Lookups over your collections, mentions and tracking agents
Requirements
OAuth or an x-api-key header; every tool requires authentication. Collecting mentions or audiences, saving a post and detecting patterns are charged in credits, and collect_mentions bills a research preview even if collection never starts. Questions to the assistant (buzzabout__ask) and natural-language queries are charged separately for model use. Lookups, status reads and rendering are free. Free trials exist, as do creator and education programmes.
Setup effort
Paste a URL, then authorize — add the endpoint to your client, then approve the OAuth consent screen
USE CASES
01
Asking what people are saying
Put the question to the chosen datasets and collect the answer asynchronously rather than waiting on one call.
02
Grounding an answer in specific posts
Reference up to 50 posts directly so the response is about those rather than the whole corpus.
03
Saving one post you found
Paste a public post's URL to save it into a collection, then check when it is ready.
15 tools
Listed in project docs. Your client may show a different list.
buzzabout__collect_audience
Profiles the authors behind a mentions collection — demographics, interests, brand affinity.
buzzabout__ingest_mention
Save one public post by its direct URL.
buzzabout__get_save_mention_status
Read the current status without starting another save.
buzzabout__detect_patterns
Clusters a collection into the themes actually present in it, each with its posts and a count.
buzzabout__query
Ask a question about collected data and get a table back.
buzzabout__get_run
Waits for a collection, audience or pattern-detection run.
buzzabout__ask
prompt (req), chat_id, dataset_ids[], post_refs[] (cap 50)
buzzabout__get_message
chat_id, message_id (both req)
buzzabout__render
block ref
buzzabout__list_collections
kind (mentions | audience), limit, cursor.
buzzabout__list_mentions
A plain filtered, sorted, cursor-paginated list of mentions.
buzzabout__list_tracking_agents
buzzabout__get_tracking_agent
buzzabout__get_me
User with account, plan, members, teams.
BEFORE YOU INSTALL
What to know first
Tools you get
15 tools named in the project's documentation.
Requires
- an MCP client that can reach a remote endpoint
- OAuth or x-api-key
INSTALLATION
Connect your client
Add the endpoint to your client, then approve the OAuth consent screen.
Server setup
ENDPOINT URL
https://mcp.buzzabout.ai/mcp/Requirements
- OAuth or x-api-key
