list_roadmapsLABSCO SUMMARY
Writes are action-based, so the write surface stays small.
Rather than one tool per create, update and delete, each object type gets a single manage tool with an action argument — twelve write tools against thirty-five read tools. That keeps the write surface reviewable and makes it obvious in a transcript when something was changed. The agent skills are the other unusual piece: the customer-facing one deliberately omits internal ids, lane names and OKRs, so a roadmap can be discussed with a customer without leaking the internals.
What it is
A Go MCP server over the ProductPlan API, shipped as a single binary. It splits its surface deliberately: 35 read tools that fetch, and 12 action-based write tools that take an action argument, so creating and editing go through a smaller, more deliberate set.
What you get
- Roadmap reads: list roadmaps, open one, and pull its bars, lanes, milestones, legends and comments — or the whole thing at once —
list_roadmaps,get_roadmap,get_roadmap_bars,get_roadmap_lanes,get_roadmap_milestones,get_roadmap_complete - Bar detail including children, comments, connections and links —
get_bar,get_bar_children,get_bar_connections,get_bar_links - OKRs: objectives and key results, read and managed —
list_objectives,get_objective,list_key_results,manage_objective,manage_key_result - Discovery: ideas, opportunities, idea forms, and the customers and tags attached to them —
list_ideas,get_idea,list_opportunities,list_all_customers,list_all_tags,manage_idea,manage_opportunity - Launches with their sections and tasks —
list_launches,get_launch_sections,get_launch_tasks,manage_launch,manage_launch_section,manage_launch_task - Roadmap writes through action-based tools covering create, update and delete —
manage_bar,manage_lane,manage_milestone,manage_bar_connection,manage_bar_link - Admin and health reads, plus users and teams —
check_status,health_check,list_users,list_teams - Pre-built agent skills for different audiences — a general workflow reference, a PM toolkit, a leadership view and a customer-facing one that omits internal ids and OKRs
- A CLI over the same binary:
productplan roadmaps,productplan objectives,productplan ideas,productplan launches
Requirements
A ProductPlan API token from Settings > API, supplied as PRODUCTPLAN_API_TOKEN — required, and it only reaches data your account can see. Download the binary for your platform from the releases page, make it executable and put it on your path, then reference it by full path in the client config; on Windows use double backslashes in JSON. A container image ghcr.io/olgasafonova/productplan-mcp-server is published at 4.8.2, and .mcpb bundles exist for macOS, Linux and Windows. Setup pages are supplied for Claude Desktop, Claude Code, Cursor, VS Code with Cline or Continue, and n8n. The token never leaves your machine — the client spawns the binary and talks over stdin and stdout.
Setup effort
Follow the published steps — chmod +x ~/Downloads/productplan-darwin-arm64
USE CASES
01
Prepping for a standup
Pull the roadmap complete, or just its bars, and summarise what changed rather than screenshotting the board.
02
Reporting on OKR progress
List objectives and their key results, and flag the ones behind schedule for a stakeholder update.
03
Triaging the idea backlog
List ideas by tag, read the ones from customers, and create or update the ones worth pursuing.
04
Coordinating a launch
Read the launch's sections and tasks to find what is still incomplete, then update the ones that are done.
05
Sharing a timeline with a customer
Use the customer-facing skill so the answer leaves out internal ids, lane names and OKRs.
47 tools
Listed in project docs. Your client may show a different list.
get_roadmapget_roadmap_barsget_roadmap_lanesget_roadmap_milestonesget_roadmap_legendsget_roadmap_commentsget_roadmap_completeget_barget_bar_childrenget_bar_commentsget_bar_connectionsget_bar_linkslist_objectivesget_objectivelist_key_resultsget_key_resultlist_ideasget_idealist_all_customerslist_all_tagslist_opportunitiesget_opportunitylist_idea_formsget_idea_formlist_launchesget_launchget_launch_sectionsget_launch_sectionget_launch_tasksget_launch_taskcheck_statushealth_checklist_userslist_teamsmanage_barmanage_lanemanage_milestonemanage_bar_connectionmanage_bar_linkmanage_objectivemanage_key_resultmanage_ideamanage_opportunitymanage_launchmanage_launch_sectionmanage_launch_taskBEFORE YOU INSTALL
What to know first
Tools you get
47 tools named in the project's documentation.
Requires
PRODUCTPLAN_API_TOKEN in your environment
INSTALLATION
Connect your client
Server setup
Download the file for your computer from the Releases page (https://github.com/olgasafonova/productplan-mcp-server/releases): productplan-darwin-arm64 (Mac M1–M4), productplan-darwin-amd64 (Mac Intel), productplan-linux-amd64 (Linux) or productplan-windows-amd64.exe (Windows).
On Mac/Linux, make it executable and move it onto your path (the filename is the one you downloaded):
TERMINAL
chmod +x ~/Downloads/productplan-darwin-arm64
sudo mv ~/Downloads/productplan-darwin-arm64 /usr/local/bin/productplanMCP Clients
Claude Desktop, Claude Code, Cline, Cursor
JSON
{
"mcpServers": {
"productplan": {
"command": "/usr/local/bin/productplan",
"env": {
"PRODUCTPLAN_API_TOKEN": "<YOUR_PRODUCTPLAN_API_TOKEN>"
}
}
}
}Continue
CONTINUE · JSON
{
"mcpServers": [
{
"name": "productplan",
"command": "/usr/local/bin/productplan",
"env": {
"PRODUCTPLAN_API_TOKEN": "<YOUR_PRODUCTPLAN_API_TOKEN>"
}
}
]
}n8n
Set this environment variable on your n8n instance:
FILE
N8N_COMMUNITY_PACKAGES_ALLOW_TOOL_USAGE=trueAdd an MCP Client node to your workflow.
Configure it with Command /usr/local/bin/productplan (Windows: C:\Tools\productplan.exe) and Environment Variables PRODUCTPLAN_API_TOKEN=<YOUR_PRODUCTPLAN_API_TOKEN>.
Connect it to an AI Agent node.
