Every tool exposed by the Octolens MCP v2 server — mentions, analytics, keywords, feeds, destinations, workspace, and usage — with what each does and when your AI assistant uses it.
The v2 server exposes 23 tools. You don’t call these directly — your AI assistant picks the right tool based on your request. This reference is for understanding what the assistant can do and for writing precise prompts.Tools marked write mutate workspace state (create or change keywords, feeds, and destinations). Everything else is read-only.
Fetch posts the pipeline collected for your workspace. Filter by source, sentiment, date range, keyword IDs, tags, language, follower counts, or engagement. Supports AND-only and advanced AND/OR filter logic, plus pagination. Returns newest-first.
analytics
Run a read-only SQL SELECT against your mentions data for aggregations, counts, trends, and breakdowns (e.g. share of voice, sentiment over time). Not for listing individual mentions.
analytics_context
Returns the analytics schema and valid values (sources, etc.). The assistant calls this before composing an analytics query.
Example prompts
“Show me the latest 20 negative mentions from Reddit and Hacker News.”
“Plot mention volume by week for the last quarter, split by sentiment.”
A keyword is a phrase (brand, competitor, product, or topic) Octolens continuously searches for across your enabled platforms. Each has a numeric id used as a filter and as the target of keyword writes.
Tool
What it does
list_keywords
List every keyword the workspace monitors, with IDs, platforms, tag, and status.
get_keyword
Fetch one keyword’s full configuration by ID.
add_keyword
Start monitoring a new phrase. Only the phrase is required — the AI fills in context, exclusions, and platforms from your company profile. Counts against your keyword limit.
update_keyword
Change a keyword’s configuration (platforms, exclusions, context, tag, etc.).
pause_keyword
Pause or resume collection for a keyword without deleting it.
delete_keyword
Permanently remove a keyword.
list_keyword_suggestions
List AI-proposed keyword tweaks pending your review.
accept_keyword_suggestion
Accept a pending keyword suggestion.
reject_keyword_suggestion
Reject a pending keyword suggestion.
Example prompts
“Add ‘vercel deployment’ as a new keyword, Reddit only.”
“Pause the keyword for our old product name.”
“Show me the pending keyword suggestions and accept the competitor ones.”
A feed is a saved filter preset (a “view”), optionally wired to one or more notification destinations — each an email list, a Slack channel set, or a webhook URL with its own cadence. Pass a feed’s id to list_mentions (as view) to reuse its filters.
Tool
What it does
list_feeds
List saved feeds with their filters and any wired-up destinations.
create_feed
Create a feed with simple (AND) or advanced (AND/OR) filters, optionally with notification destinations.
update_feed
Change a feed’s name, icon, or filters.
delete_feed
Permanently delete a feed.
add_feed_destination
Attach an email, Slack, or webhook destination (with its own cadence) to a feed.
remove_feed_destination
Remove a destination from a feed.
search_slack_channels
Resolve a Slack channel name to the channel ID that Slack destinations require.
Slack destinations need channel IDs, not names — the assistant calls search_slack_channels first to resolve them. Slack must be connected to your workspace, or writes are rejected.
Example prompts
“Create a ‘Buy intent on Reddit’ feed and send a daily 9am digest to #leads in Slack.”
“Add a webhook destination to my Competitor Activity feed pointing at https://example.com/hook.”
Return workspace context — name, plan, enabled platforms — and the monitored company profile (industry, description, competitors, relevance guidelines). The assistant calls this once at the start of a session to ground tone and disambiguate mentions.
get_usage
Return current usage against your plan limits (e.g. keyword and mention headroom).
get_global_filters
Return the workspace-wide global filters applied across all queries.
list_tags
List valid tag values (AI-assigned topical labels like competitor_mention, buy_intent, bug_report) for use in mention filters.
Example prompts
“What plan am I on and how many keywords can I still add?”
“What tags are available, and how many bug reports came in this week?”
