Competiflow MCP Server
Competiflow MCP server lets MCP-compatible AI clients manage competitor monitoring via the Competiflow API.
List and create workspaces.
Add, list, read, update, and delete competitors, including polling discovery status.
List, read, create, update, and delete monitors; set status, cadence, type, URL, and min severity.
Queue on-demand checks and refresh all active monitors or one competitor.
Query the change feed with pagination and filters for time, severity, monitor type, competitor, and review status.
Read full changes, acknowledge/dismiss them, and request evidence payloads (capture, markdown, raw diff).
Fetch workspace digests.
Use prompts/resources for weekly briefings, triage, competitor tracking, and an agent guide/digest.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@Competiflow MCP Server Show me recent changes from my competitors."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Competiflow MCP Server
MCP (Model Context Protocol) server for Competiflow. Use competitor monitoring as tools in Cursor, Claude Desktop, Windsurf, and any MCP-compatible client.
Track competitors, run discovery, read the interpreted change feed, triage items, and pull workspace digests without leaving your AI client.
Setup
Create an API key in the Competiflow dashboard under API keys, then add the server to your client config. COMPETIFLOW_API_KEY is the only variable you need.
Cursor
Add to ~/.cursor/mcp.json:
{
"mcpServers": {
"competiflow": {
"command": "npx",
"args": ["-y", "@competiflow/mcp-server"],
"env": {
"COMPETIFLOW_API_KEY": "sk_live_your_key_here"
}
}
}
}Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"competiflow": {
"command": "npx",
"args": ["-y", "@competiflow/mcp-server"],
"env": {
"COMPETIFLOW_API_KEY": "sk_live_your_key_here"
}
}
}
}Windsurf
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"competiflow": {
"command": "npx",
"args": ["-y", "@competiflow/mcp-server"],
"env": {
"COMPETIFLOW_API_KEY": "sk_live_your_key_here"
}
}
}
}Related MCP server: Territory News MCP Server
Available Tools
Tool | API | Description |
|
| List workspaces with review counts |
|
| Create a workspace (optional |
|
| Rename, set or clear |
|
| Remove a workspace and all its history |
|
| Add a competitor by homepage URL. Async discovery, idempotent on duplicate URL |
|
| List competitors with monitors and |
|
| Read one competitor |
|
| Set cadence or status on all its monitors |
|
| Remove a competitor and all its history |
|
| List monitors with status, cadence, and |
|
| Read one monitor. Pass |
|
| Add a manual monitor on a specific URL |
|
| Update URL, cadence, status, or |
|
| Remove a monitor and all its history |
|
| Queue an on-demand check |
|
| Queue checks for active monitors |
|
| Change feed with |
|
| Full change. Pass |
|
| Set |
|
| Weekly workspace summary |
Every tool declares MCP annotations (readOnlyHint, destructiveHint, idempotentHint, openWorldHint), so clients can ask before running the delete_* tools.
Prompts
Prompt | Arguments | What it does |
|
| Summarizes competitor activity for a period, most important first |
|
| Walks the unreviewed inbox and proposes acknowledge or dismiss, then asks before updating |
|
| Adds a competitor, waits for discovery, and reports what is monitored |
Without workspace_id, each prompt starts with list_workspaces.
Resources
URI | Type | Contents |
|
| Agent guide: the typical flow, discovery, evidence, and delete versus pause |
|
| 7-day digest of one workspace. Clients that list resources show one per workspace. |
Typical flow
list_workspacesorcreate_workspace(optionalproduct_urlimproves competitive context on changes).add_competitorwith a homepage URL.get_competitoruntildiscovery_statusiscompleted.refresh_workspaceorcheck_monitorto pull fresh data.list_changeswithsince=7dandmin_severity, orget_digestfor a summary.get_changewithincludewhen you need evidence.update_changeto acknowledge or dismiss items.
Discovery and idempotency
add_competitor starts discovery in the background and returns an acknowledgement, not the finished result. Poll get_competitor until discovery_status is completed.
A workspace runs one discovery at a time. Posting the same homepage URL again returns the existing competitor without starting a second run. If a different competitor is still discovering, the call returns 409 discovery_in_progress with active_competitor_id so you know which one to poll.
Evidence with include
get_change and get_monitor accept an include array to attach the underlying capture in one call:
Token | Available on | Returns |
|
| Extracted fields from the snapshot |
|
| Page markdown from the snapshot |
|
| Before and after diff for the change |
For a change the capture is the post-change page. For a monitor it is the latest successful check, which is the current baseline.
Delete versus pause
delete_monitor, delete_competitor, and delete_workspace are permanent. They remove the record along with its check runs, snapshots, and changes. To stop checks while keeping intel history, call update_monitor, update_competitor, or update_workspace with status: paused instead.
delete_workspace returns 409 discovery_in_progress while a competitor in the workspace is still being discovered.
Agent guidelines (for AI clients)
Poll discovery, do not assume: after
add_competitor, pollget_competitoruntildiscovery_statusiscompletedbefore reading changes. On409, poll theactive_competitor_idfrom the error.Filter the feed: pass
since(1d,7d,30d, or ISO8601),min_severity,monitor_type(homepage,blog,docs,changelog,pricing), andreview_statustolist_changesinstead of pulling everything. Usemeta.countsfor inbox totals andnext_cursorto page.Interpretation fields: changes and digest items expose
why_it_mattersfor competitive context.recommended_actionwas removed from the API.Fetch evidence on demand: only add
includetoget_changeorget_monitorwhen you need the underlying capture. It adds payload.Prefer pause over delete: default to
status: pausedwhen a user wants to stop a monitor. Reservedelete_*for permanent removal, since it drops history.
How it works
The server translates each tool call into a Competiflow v1 API request and returns structured JSON. Your agent sees interpreted changes, severity, and why it matters instead of raw page scrapes.
AI Agent → MCP Tool Call → Competiflow API → Interpreted Change Feed → Structured Response → AI AgentEnvironment variables
Variable | Required | Description |
| Yes | Your Competiflow API key ( |
Get an API key
Sign up at competiflow.com and create a key under API keys.
Development
npm install
npm run build
COMPETIFLOW_API_KEY=sk_live_... node dist/index.jsnpm test builds and runs a smoke test that starts the server without a key, the way directory scanners do, and checks the tools, annotations, prompts, and resources.
License
MIT
Available Tools
20 toolsadd_competitorAdd competitorAIdempotent
Add a competitor to a workspace by homepage URL and start discovering its pricing, docs, blog, and changelog pages. Discovery is async: poll get_competitor until discovery_status is completed. Re-adding the same URL returns the existing competitor. Returns 409 with active_competitor_id while another discovery runs in the workspace. To watch one specific page, use create_monitor.
Inputs: url (required), workspace_id (required), monitor_types.
check_monitorCheck monitor nowA
Queue an immediate check of one monitor. Uses one of your monthly checks and returns 429 refresh_throttled if the monitor was checked too recently. Results arrive asynchronously: read them later with list_changes or get_monitor. To check every monitor in a workspace, use refresh_workspace.
Inputs: monitor_id (required).
create_monitorCreate monitorA
Watch one specific page of an existing competitor, such as a pricing or docs URL that discovery missed. To add a new competitor and discover its pages automatically, use add_competitor.
Inputs: url (required), monitor_type (required), competitor_id (required), cadence.
create_workspaceCreate workspaceA
Create a workspace for one product or market. Set product_url to your own product so why_it_matters explains each change relative to it. Returns 422 when your plan's workspace limit is reached. To rename or change an existing workspace, use update_workspace.
Inputs: name (required), product_url.
delete_competitorDelete competitorADestructiveIdempotent
Permanently delete a competitor with its monitors, snapshots, and changes. Cannot be undone. To stop checks and keep history, use update_competitor with status paused.
Inputs: competitor_id (required).
delete_monitorDelete monitorADestructiveIdempotent
Permanently delete one monitor with its snapshots and changes. Cannot be undone. To stop checks and keep history, use update_monitor with status paused.
Inputs: monitor_id (required).
delete_workspaceDelete workspaceADestructiveIdempotent
Permanently delete a workspace with all its competitors, monitors, snapshots, and changes. Cannot be undone. Returns 409 discovery_in_progress while a competitor is still being discovered. To stop checks and keep history, use update_workspace with status paused.
Inputs: workspace_id (required).
get_changeGet changeARead-only
Get one change with why_it_matters, field-level diffs, and related clusters. Set include to add the captured page or the raw diff as evidence.
Inputs: change_id (required), include.
get_competitorGet competitorARead-only
Get one competitor with its monitors and discovery fields. Use it to poll discovery after add_competitor until discovery_status is completed or failed.
Inputs: competitor_id (required).
get_digestGet workspace digestARead-only
Get a workspace summary for a period: changes needing review with why_it_matters, an activity map by competitor, and top opportunities. Use it for briefings; to page through individual changes, use list_changes.
Inputs: workspace_id (required), period.
get_monitorGet monitorARead-only
Get one monitor's settings, health, and last run. Set include to also return the latest successful capture as structured data or markdown.
Inputs: monitor_id (required), include.
list_changesList changesARead-only
List detected competitor changes in a workspace, newest first, with summary, severity, why_it_matters, monitor_type, and review_status. meta.counts gives inbox totals; page with after set to next_cursor. For a summarized period, use get_digest; for evidence on one change, use get_change.
Inputs: workspace_id (required), after, limit, since, +4 more.
list_competitorsList competitorsARead-only
List the competitors in a workspace with discovery_status and their monitors, each with last_run. Use it for an overview of what is tracked. To list monitors alone or filter them by competitor, use list_monitors.
Inputs: workspace_id (required).
list_monitorsList monitorsARead-only
List the monitors in a workspace, optionally for one competitor, with URL, type, cadence, status, health, and last_run. Use it to find a monitor_id or spot failing monitors. For competitors with their monitors nested, use list_competitors.
Inputs: workspace_id (required), competitor_id.
list_workspacesList workspacesARead-only
List all your workspaces with competitor and monitor counts, needs_review_count, and the highest unreviewed severity. Start here to find a workspace_id. For one workspace's recent activity, use get_digest.
Inputs: no named parameters.
refresh_workspaceRefresh workspaceA
Queue immediate checks for every active monitor in a workspace, or for one competitor when competitor_id is set. Each monitor checked uses one monthly check; recently checked monitors are skipped and listed in skipped_ids. Returns 409 while discovery is running. For a single monitor, use check_monitor.
Inputs: workspace_id (required), competitor_id.
update_changeTriage changeAIdempotent
Triage a change by setting its review_status: acknowledged when it is relevant, dismissed when it is noise, or unreviewed to put it back in the inbox.
Inputs: change_id (required), review_status (required).
update_competitorUpdate competitorAIdempotent
Set the check cadence or pause/resume every monitor of one competitor. History is kept. To change a single monitor, use update_monitor; for a whole workspace, use update_workspace.
Inputs: competitor_id (required), status, cadence.
update_monitorUpdate monitorAIdempotent
Change one monitor's URL, type, cadence, status, or min_severity. Use status paused to stop checks and keep history. To change every monitor of a competitor at once, use update_competitor.
Inputs: monitor_id (required), url, status, cadence, +2 more.
update_workspaceUpdate workspaceAIdempotent
Rename a workspace, set or clear its product_url, or pause/resume every monitor in it. Send only the fields to change; nothing is written if any field is invalid. Pausing keeps all history, so prefer it over delete_workspace. To pause one competitor, use update_competitor.
Inputs: workspace_id (required), name, status, product_url.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 20 tools
Each tool maps to a distinct resource+action pair, and the descriptions actively disambiguate the closest neighbours (check_monitor vs refresh_workspace, list_competitors vs list_monitors, list_changes vs get_digest vs get_change, add_competitor vs create_monitor) with explicit 'use X instead' pointers. Pause-vs-delete semantics are also spelled out consistently across the three hierarchies, so an agent can select confidently.
The set follows a predictable verb_noun pattern (list_/get_/create_/update_/delete_ + resource, plus check_/refresh_ actions), which reads cleanly. The only wrinkle is that creation uses two verbs — add_competitor alongside create_monitor/create_workspace — a minor inconsistency rather than a real hazard.
Twenty tools is on the heavy side of the rubric, but it decomposes cleanly: near-complete CRUD across three resources (workspace, competitor, monitor) plus four change/digest tools and two on-demand check tools. Each tool earns its place; nothing looks like a redundant duplicate.
The surface covers full lifecycle management of workspaces, competitors, and monitors, plus change reading (list_changes, get_change), triage (update_change), and briefing (get_digest) with async discovery/refresh flows handled via polling and explicit status codes. Minor gaps remain: there is no get_workspace for a single workspace by id (only list_workspaces), and change triage is one-at-a-time with no bulk operation.
Maintenance
Related MCP Connectors
Track competitors from your AI assistant: change feed, page snapshots, labels and alerts.
Monitor competitors, build research corpora or give AI agents structured metadata.
Live context for your agent: competitors, mentions and markets, from public data.
Competitive intelligence MCP server for Claude and ChatGPT. Ripplewatch tracks your competitors' pricing, hiring, product launches, press and reviews, gives each one a Momentum score from -100 to +100, and checks every move against your own won and lost deals. Ask what changed this week, which competitor is heating up, or what is costing you deals, and log a deal just by telling it. Works in any assistant that supports remote MCP.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceAutonomous competitive intelligence tracking competitors across LinkedIn, news, reviews, job postings, and regulatory signals, generating executive briefs and sales battlecards.PolyForm Noncommercial 1.0.0
- AlicenseAqualityDmaintenanceProvides real-time account intelligence, deal signals, and strategic openers to AI agents and MCP-compatible orchestrators, enabling automated sales workflows such as pre-call battlecards, deal reactivation, and territory monitoring.12MIT
- AlicenseNot gradedqualityCmaintenanceEnables real-time web page fetching, web searches, RSS/Atom feed parsing, and competitor mention tracking to monitor market and competitor changes.MIT
- AlicenseAqualityBmaintenanceEnables AI assistants such as Claude, Cursor, and Codex to pull structured, source-backed, dated competitive intelligence — positioning, pricing, ads, social, reviews, hiring, momentum, and exploitable weaknesses — for any company in the universe, alongside the user's own projects, reports, battlecards, timelines, and evidence. Read-only by default, with optional opt-in competitor creation and plan-gated tools for deeper analysis.16108 npmMIT