pension-pro-mcp
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., "@pension-pro-mcp search for plans with status 'active' and type '401k'"
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.
PensionPro MCP Server
A local, open-source MCP server that provides AI assistants with tools to interact with the PensionPro REST API.
Features
Plan Lookup & Search — Search plans by name, status, type, or client. Get comprehensive plan details with contacts, cycles, and fee schedules.
Project & Task Workflow — Search projects, view task details, complete/uncomplete tasks, reassign tasks, and create projects from templates.
Client & Contact Lookup — Search clients and contacts, view client details with associated plans.
To-Do Management — Search, create, and update to-dos linked to plans, projects, or contacts.
Notes — Add and retrieve notes on plans, projects, tasks, and contacts.
Related MCP server: Jilebi
Prerequisites
Python 3.12+
A PensionPro API key and username
Installation
Install from PyPI:
pip install pension-pro-mcpOr run directly with uvx (no install required):
uvx pension-pro-mcpConfiguration
Set the following environment variables:
export PENSION_PRO_API_KEY=your_api_key
export PENSION_PRO_USERNAME=your_usernameClaude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"pension-pro": {
"command": "uvx",
"args": ["pension-pro-mcp"],
"env": {
"PENSION_PRO_API_KEY": "your_api_key",
"PENSION_PRO_USERNAME": "your_username"
}
}
}
}Claude Code
Add to your project's .mcp.json:
{
"mcpServers": {
"pension-pro": {
"command": "uvx",
"args": ["pension-pro-mcp"],
"env": {
"PENSION_PRO_API_KEY": "your_api_key",
"PENSION_PRO_USERNAME": "your_username"
}
}
}
}Available Tools
Tool | Description |
| Search and filter plans by name, status, type, or client |
| Get plan with contacts, cycles, services, investments, fees |
| Get projects for a plan with task completion summaries |
| Search and filter projects by status, type, or plan |
| Get project with task groups, tasks, participants, notes, and files |
| Get a single task with state, assignment, and notes |
| Mark a task as complete |
| Revert a task to incomplete |
| Reassign a task to a different employee |
| Create a new project from a template |
| Search and filter clients by company name |
| Get client with plans and notes |
| Search and filter contacts by name or client |
| Search and filter to-dos |
| Get a to-do with its comments |
| Create a new to-do linked to an entity |
| Update a to-do's details |
| Add a note to a plan, project, task, or contact |
| Get notes for an entity |
| Search PensionPro API endpoints by keyword |
| Get full details for a specific API endpoint |
| Search API data models/schemas by keyword |
| Get the full definition of an API data model |
| Search PensionPro help center articles by keyword |
| Get the full content of a help article |
| List available help sections with article counts |
Development
git clone https://github.com/douglaslinsmeyer/pension-pro-mcp.git
cd pension-pro-mcp
pip install -e ".[dev]"
pytestRefreshing Help Articles
The bundled help articles can be refreshed from the PensionPro knowledge base:
python scripts/scrape_docs.pyLicense
MIT
Available Tools
32 toolsadd_noteA
Add a note to a plan, project, task, or contact. At least one entity ID must be provided.
Inputs: text (required), plan_id, project_id, task_id, +2 more.
complete_taskC
Mark a task as complete.
Inputs: task_id (required).
create_project_from_templateC
Create a new project on a plan from a project template.
Inputs: plan_id (required), template_id (required).
create_todoB
Create a new to-do, optionally linked to a plan, project, or contact.
Inputs: subject (required), description, priority_id, due_date, +5 more.
get_api_endpointB
Get details for a specific API endpoint. Returns a compact summary by default. Set raw=true for the full JSON schema.
Inputs: path (required), raw. Returns: result
get_api_schemaA
Get the definition of a specific API data model/schema. Returns a compact summary by default. Set raw=true for the full JSON schema.
Inputs: name (required), raw. Returns: result
get_client_detailsC
Get a client with their plans and notes.
Inputs: client_id (required).
get_employee_statsA
Get individual employee performance analysis across all worktrays.
Searches by last name, then computes per-worktray metrics for workload, throughput, and quality (rejections and bounce-backs). Returns a compact summary; full results cached and available via the employee-stats resource.
If the name matches multiple employees, returns a candidate list to disambiguate.
Inputs: name (required), days_back.
get_help_articleB
Get the full content of a PensionPro help article by its ID.
Inputs: article_id (required).
get_notesA
Get notes for an entity (plan, project, task, or contact). At least one entity ID must be provided.
Inputs: plan_id, project_id, task_id, contact_id, +1 more. Returns: result
get_plan_detailsB
Get comprehensive details for a single plan.
Returns the plan record with contacts, cycles, services, investments, and fee schedules.
Inputs: plan_id (required).
get_plan_projectsB
Get all projects for a plan with task completion summaries.
Each project includes a task_summary with total, completed, and pending counts.
Inputs: plan_id (required), status. Returns: result
get_project_detailsB
Get a project with its task groups, tasks, participants, notes, and files.
Inputs: project_id (required).
get_task_detailsA
Get a single task with its current state, assignment, and notes.
Inputs: task_id (required).
get_task_groupB
Get a task group with all its tasks.
Inputs: task_group_id (required).
get_task_group_cycle_timesA
Compute task group cycle times segmented by project template.
Analyzes completed task groups within the lookback window to reveal which workflows are slow and which steps are bottlenecks. Returns per-template stats including avg/median/min/max cycle time and SLA adherence.
Set include_steps=true to fetch per-task step durations and identify the bottleneck step within each template. This makes additional API calls (one per task group) and may be slow for large result sets.
Inputs: days_back, plan_id, template_id, include_steps.
get_todoB
Get a to-do with its comments.
Inputs: todo_id (required).
get_worktrayB
Get a worktray with its members and all active (incomplete) tasks routed to it.
Inputs: worktray_id (required).
get_worktray_member_statsA
Get per-member workload, performance, and queue health metrics for a worktray.
Analyzes completed tasks within the lookback window and current active tasks. Returns a compact summary with the top 20 members by activity, aggregate stats, and queue health. Full results (all members, individual task lists, overdue task details) are written to a cache file and available via the worktray-stats resource.
Inputs: worktray_id (required), days_back.
get_worktraysB
List all worktrays. Set active_only=false to include inactive worktrays.
Inputs: active_only, limit. Returns: result
list_help_sectionsA
List all available PensionPro help center sections with article counts.
Inputs: no named parameters. Returns: result
reassign_taskC
Reassign a task to a different employee.
Inputs: task_id (required), assigned_to_id (required).
search_api_pathsA
Search PensionPro API endpoints by keyword. Use this to discover available API paths.
Inputs: keyword (required). Returns: result
search_api_schemasA
Search PensionPro API data models by keyword. Use this to discover field names and types.
Inputs: keyword (required). Returns: result
search_clientsB
Search and filter clients by company name.
Inputs: name, limit. Returns: result
search_contactsC
Search and filter contacts by name or client.
Inputs: name, client_id, limit. Returns: result
search_help_articlesA
Search PensionPro help center articles by keyword. Optionally filter by section name. Use list_help_sections to see available sections.
Inputs: keyword (required), section. Returns: result
search_plansA
Search and filter pension plans by name, status, type, or client.
Returns a summary list of matching plans.
Inputs: name, status, plan_type, client_name, +1 more. Returns: result
search_projectsC
Search and filter projects by status, type, or plan.
Inputs: status, project_type, plan_id, limit. Returns: result
search_todosC
Search and filter to-dos by status, priority, plan, or project.
Inputs: status, priority, plan_id, project_id, +1 more. Returns: result
uncomplete_taskB
Undo task completion — revert a task to incomplete.
Inputs: task_id (required).
update_todoC
Update an existing to-do's subject, description, status, priority, or due date.
Inputs: todo_id (required), subject, description, status_id, +2 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 32 tools
Each tool targets a distinct entity or operation; search, get, create, and action tools are clearly separated with no overlapping purposes.
All tool names follow a consistent verb_noun pattern (e.g., add_note, get_plan_details, search_plans) with only minor deviations like uncomplete_task, which still fits the pattern.
32 tools is excessive for a typical MCP server; while the domain is broad, the count overwhelms the coherence and suggests potential redundancy (e.g., multiple API schema tools).
Notable gaps exist: no create/update/delete for plans, projects, or tasks (only create_todo and complete_task exist). The surface is incomplete for full lifecycle management.
Maintenance
Related MCP Connectors
MCP server that lets AI assistants use all OneSchema features exposed via the public API.
MCP server for OpenAI API (chat completions, image generation, embeddings) via AceDataCloud
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
MCP server that lets your AI assistant build and deploy apps, hosting each one at a live URL.
Related MCP Servers
- AlicenseBqualityDmaintenanceA personal MCP server for AI assistant integration that provides custom tools, resources, and prompts for use with Claude Desktop and other MCP-compatible clients.2MIT
- AlicenseNot gradedqualityCmaintenanceA plugin-based MCP server that enables AI assistants to interact with external systems through custom tools, resources, and prompts.4AGPL 3.0
- AlicenseNot gradedqualityCmaintenanceA self-hostable MCP server that turns a folder of skills into callable tools via MCP and REST APIs.3MIT
- AlicenseNot gradedqualityCmaintenanceA generic MCP server that converts any OpenAPI/Swagger specification into MCP tools, enabling AI assistants to search, explore, and execute REST APIs.MIT