Tools Reference

The MCP server exposes tools across mentions, listeners, search, feeds, notifications, webhooks, analytics, and workspace. All tools operate on your workspace and respect your plan limits.

If you are unfamiliar with terms like listener, query, fit score, or intent tags, see the Concepts page first.

Fetches brand mentions from your feed with rich filtering. You can filter by platform, date range, sentiment, AI fit score, intent tags, keywords, subreddit, and seen/unseen status. Results are paginated at 50 per page.

Use when: You want to read, summarize, or triage recent mentions. Pair filters with snitchfeed_list_listeners to scope to a specific topic.

Listeners

List Listeners

snitchfeed_list_listeners

Returns all active and paused listeners in your workspace, including each listener's ID, name, status, intent description, platform configuration, and any tracked authors (Tracked Profiles).

Use when: You need listener IDs to pass into other tools, or want to review what your workspace is currently monitoring.

List Queries

snitchfeed_list_queries

Lists all keyword queries across your listeners. Each query includes the raw query text, which platforms it runs on, and any platform-specific filters (subreddits, LinkedIn author filters, Twitter minimum likes).

Use when: You want to audit or modify specific keyword rules, or filter mentions by the query that triggered them.

Create Listener

snitchfeed_create_listener

Creates a new listener with keyword queries and/or tracked LinkedIn/X profiles. Provide at least one query string (boolean syntax: AND, OR, NOT, parentheses) or at least one tracked_authors entry (after resolving each URL with Resolve Tracked Author Profile), or both. Optionally include an intent description that tunes the AI fit score.

Use when: You want to start monitoring a new brand, competitor, topic, or specific LinkedIn/X accounts from scratch.

Update Listener

snitchfeed_update_listener

Updates a listener's name, intent description, status (active or paused), or color. Queries are not affected. Supports adding tracked_authors (after resolving each profile URL) and removing them via remove_tracked_authors (by platform + profile URL, never by internal ID).

Use when: You want to rename a listener, update what the AI considers "relevant" for fit scoring, pause monitoring temporarily, or change which LinkedIn/X accounts the listener follows as tracked profiles.

Resolve Tracked Author Profile

snitchfeed_resolve_tracked_author_profile

Resolves a pasted LinkedIn or Twitter/X profile URL to the platform account it points to (display name, platform, profile type, and related fields). Confirm the resolved account with the user, then pass the tool output unchanged into create or update listener as tracked_authors.

Use when: The user wants to follow a specific person or company page on a listener, not just keyword-match posts. See Tracked Author and the product guide Tracked Profiles.

Create Listener Query

snitchfeed_create_listener_query

Adds a new keyword query to an existing listener. Triggers a historical backfill for the new query automatically.

Use when: You want to expand an existing listener to watch for additional keywords without creating a new listener.

Update Listener Query

snitchfeed_update_listener_query

Updates the query text, platform scope, or filter settings of an existing keyword query. If the query text changes, irrelevant existing mentions may be archived and new keywords are backfilled.

Use when: You want to refine a keyword rule, change which platforms it runs on, or adjust LinkedIn/Twitter filters.

Delete Listener

snitchfeed_delete_listener

Permanently deletes a listener and all its keyword queries. The AI client will always ask for explicit confirmation before calling this tool.

Use when: You want to remove a listener and stop monitoring its topic entirely. This cannot be undone, though historical mentions are preserved even after the listener and its queries are gone.

Delete Listener Query

snitchfeed_delete_listener_query

Permanently removes a single keyword query from a listener. The AI client will always ask for explicit confirmation before calling this tool. A listener must retain at least one keyword query or tracked author, so deleting the last query is only allowed when tracked profiles remain; otherwise delete the listener itself.

Use when: You want to narrow an existing listener's scope by removing one of its keyword rules without affecting the others.

Get Query Grammar

snitchfeed_get_query_grammar

Returns the full boolean query syntax reference: supported operators, term limits, and platform-specific constraints.

Use when: You are constructing a query string and need to check the syntax or limits before passing it to a create or update tool.

Search LinkedIn Industries

snitchfeed_search_linkedin_industries

Searches LinkedIn industry codes by name or category, returning each match's numeric ID, label, and hierarchy path. Omit the search term to browse all top-level industry groups (results capped at 25).

Use when: You want to scope a LinkedIn query to specific industries. Pass the returned id as linkedin_author_industry when creating or updating a listener query.

Test Twitter Query

snitchfeed_test_twitter_query

Validates a Twitter/X keyword query and runs a live search to preview matching posts before you save it, with optional language and minimum-likes filters. Free — testing does not cost credits.

Use when: Required before saving any query with Twitter enabled, via snitchfeed_create_listener, snitchfeed_create_listener_query, or snitchfeed_update_listener_query. Review the sample posts and confirm relevance with the user before saving.

Test LinkedIn Query

snitchfeed_test_linkedin_query

Validates a LinkedIn keyword query and runs a live search to preview matching posts, with optional job title, company, industry, and post-type filters. Free — testing does not cost credits. Comma-separated job titles run one search per title (max 3); that same multiplier applies to the recurring poll cost (4 credits per job-title search, each interval) once the query is saved.

Use when: Required before saving any query with LinkedIn enabled. Industry filtering is strongly recommended: look up numeric IDs with snitchfeed_search_linkedin_industries first.

Test HackerNews Query

snitchfeed_test_hackernews_query

Validates a Hacker News keyword query and runs a live search (via Algolia's HN index) to preview matching stories and comments. Supports boolean syntax plus HN qualifiers (author:/by:, story:, points>/points<, comments>/comments<, exact phrases, and -exclude). Free — testing does not cost credits.

Use when: Required before saving any query with Hacker News enabled. Review sample posts and confirm relevance with the user before saving.

Search

Search Tweets

snitchfeed_search_tweets

Ad hoc live search of Twitter/X for posts matching a keyword query, with optional language and minimum-likes filters. Nothing is saved. Costs 2 poll credits per call. Supports cursor-based pagination (cursor in, next_cursor/has_more out) to fetch more of the same query. The AI client is instructed to assess the relevance of each batch of results, page for more of the same when quality is good but volume is low, and revise the query instead of paging when results are off-topic.

Use when: You want to explore what's being said about a topic, brand, or competitor right now, without creating a listener.

Search LinkedIn Posts

snitchfeed_search_linkedin_posts

Ad hoc live search of LinkedIn for posts matching a keyword query, with optional job title, company, industry, and post-type filters. Nothing is saved. Costs 4 poll credits per job-title search. Supports page-based pagination (page) to fetch more of the same query. The AI client is instructed to assess the relevance of each batch of results, increment page for more of the same when quality is good but volume is low, and revise the query (resetting page to 1) when results are off-topic.

Use when: You want to explore what's being said about a topic on LinkedIn, without creating a listener. Industry filtering is strongly recommended: look up numeric IDs with snitchfeed_search_linkedin_industries first.

Search HackerNews Posts

snitchfeed_search_hn_posts

Ad hoc live search of Hacker News for stories and comments matching a keyword query. Nothing is saved. Costs 2 poll credits per call. Supports the same boolean syntax and HN qualifiers as Test HackerNews Query, plus optional story/comment scope and page-based pagination. The AI client is instructed to assess relevance per batch, page for more when quality is good, and revise the query when results are off-topic.

Use when: You want to explore Hacker News discussion on a topic without creating a listener.

Extract LinkedIn Post Comments

snitchfeed_extract_linkedin_post_comments

Scrapes commenters from a public LinkedIn post into a downloadable CSV. Costs 1 credit per comment scraped. Requires post_url and max_comments (100-1000); optional provider (harvest or rapidapi). The client should ask you for the post URL and count before calling. Returns download_url for this run's CSV and links.past_runs for the platform tool page (Past runs tab).

Use when: You want commenter profiles or comment text for a specific LinkedIn post, not keyword search (use Search LinkedIn Posts for that).

Extract LinkedIn Post Reactions

snitchfeed_extract_linkedin_post_reactions

Scrapes people who reacted to a public LinkedIn post into a downloadable CSV. Costs 1 credit per reaction scraped. Requires post_url and max_reactions (100-1000); optional provider (harvest or rapidapi). The client should ask you for the post URL and count before calling. Returns download_url for this run's CSV and links.past_runs for the platform tool page (Past runs tab).

Use when: You want reactor profiles or reaction types for a specific LinkedIn post, not keyword search (use Search LinkedIn Posts for that).

List LinkedIn Comments Scrapes

snitchfeed_list_linkedin_comments_scrapes

Lists past LinkedIn comment scrape runs for your workspace (newest first). Each run includes the post URL, comment count, download_url for the CSV, and links.past_runs for the same history on the platform.

Use when: You want to find or re-download a previous comment scrape.

List LinkedIn Reactions Scrapes

snitchfeed_list_linkedin_reactions_scrapes

Lists past LinkedIn reaction scrape runs for your workspace (newest first). Each run includes the post URL, reaction count, download_url for the CSV, and links.past_runs for the same history on the platform.

Use when: You want to find or re-download a previous reaction scrape.

Feeds

List Feeds

snitchfeed_list_feeds

Lists all saved feeds (filtered views) in your workspace. Each feed includes its filters, a dashboard url, and a notifications object (enabled status, Slack channel names only, Discord webhook URLs, email digest settings, and org webhook destinations). Top-level slack_connected and org_webhooks help configure notification tools. Slack channel ids are never included in this response.

Use when: You need feed IDs/names, current alert routing, or available org webhooks before updating notifications.

Create Feed

snitchfeed_create_feed

Creates a saved feed, which is a named filtered view of your mentions. You can scope it to specific listeners, platforms, sentiment, fit score buckets, intent tags, keywords, subreddits, date range, and seen status.

Use when: You want to build a persistent view in the SnitchFeed dashboard, such as "High-fit Buyer Signals" or "Competitor Complaints."

Update Feed

snitchfeed_update_feed

Updates a saved feed's name, sort order, or any of its filter settings. Providing a filter array fully replaces the existing value for that field. Does not change Slack/Discord/email/webhook destinations; use snitchfeed_update_feed_notifications for that.

Use when: You want to adjust the scope of an existing saved view without recreating it.

Update Feed Notifications

snitchfeed_update_feed_notifications

Updates notification destinations for a saved feed: Slack channels, Discord webhook URLs, email digests, and/or org webhooks, plus enable/disable. Partial patch: omit a field to leave it unchanged; pass an empty array to clear that destination type; pass email: null to remove email digests. Multiple Slack channels per feed are supported. Use snitchfeed_get_slack_channels before setting Slack destinations. Newly added Slack channels are joined by the bot when possible.

Use when: You want to route feed matches to Slack, Discord, email, or webhooks, or change existing alert destinations.

Get Slack Channels

snitchfeed_get_slack_channels

Lists public Slack channels in the connected workspace (id and name). Private channels are not included. When talking to the user, only show name as #name; never show Slack channel ids. If Slack is not connected, returns an error with a link to the Slack integrations page so an admin can click Connect Slack.

Use when: You need channel IDs/names before calling snitchfeed_update_feed_notifications with Slack destinations.

Delete Feed

snitchfeed_delete_feed

Permanently deletes a saved feed. Any notification routes attached to it stop routing, but existing mentions are unaffected. The AI client will always ask for explicit confirmation before calling this tool.

Use when: You want to remove a saved view that is no longer useful. This cannot be undone.

Analytics

Query Analytics

snitchfeed_query_analytics

Runs an ad-hoc analytics query against your mention data. You choose a metric (mention count, unique authors, average sentiment, average fit score), a grouping dimension (day, platform, sentiment, fit score, keyword, subreddit, language, or listener), and an optional second breakdown dimension. Results include the resolved date range.

Use when: You want a one-off data question answered: "Which platform drove the most mentions last month?" or "How is sentiment trending day by day?"

List Analytics Reports

snitchfeed_list_analytics_reports

Lists all saved analytics reports in your workspace, including each report's ID, name, a human-readable description of its query, and when its cached data was last refreshed.

Use when: You want to find an existing report before creating a new one (to avoid duplicates), or to get a report ID for updating.

Create Analytics Report

snitchfeed_create_analytics_report

Saves a named analytics query as a reusable report visible in the SnitchFeed dashboard. The query runs immediately on creation and results are cached for 15 minutes.

Use when: You want a recurring view of a specific analytics breakdown to persist in the dashboard across sessions.

Update Analytics Report

snitchfeed_update_analytics_report

Updates a saved report's name or query configuration. If any query parameter changes, unspecified parameters reset to their defaults. If only the name changes, the existing query is preserved.

Use when: You want to rename a report or change what it measures without deleting and recreating it.

Audit Query Noise

snitchfeed_audit_query_noise

Audits saved queries for noise and cost efficiency in a single call. Computes estimated monthly credit cost, poll efficiency (dead polls that returned zero new mentions), and mention quality (fit score distribution, exclusion-archive rate, thumbs-up/down feedback), then combines them into a noise score and a suggested action: keep, narrow, increase_interval, or kill.

Use when: You want to find noisy or wasteful keyword queries and decide what to fix. It does not rewrite queries itself; pair it with Get Query Grammar and Update Listener Query to apply changes.

Workspace

Get Concepts

snitchfeed_get_concepts

Returns a full explanation of SnitchFeed concepts: listeners, queries, mentions, fit score, intent, negative intent, intent tags, sentiment, platforms, analytics, and analytics reports.

Use when: You are unfamiliar with SnitchFeed terminology, or an AI client needs orientation before helping with a complex task. The same content is available in the Concepts page.

Get Organization Context

snitchfeed_get_org_context

Returns your organization's brand context: company name, website, product description, positioning, target audience, use cases, competitors, and differentiators.

Use when: You want the AI to understand your brand before interpreting mention relevance, drafting reply suggestions, or generating a report.

Get Usage

snitchfeed_get_usage

Returns current billing period usage against your plan limits: keywords used, listener count, and AI processing credits consumed.

Use when: You want to check headroom before creating new listeners or queries.

On this page