About summitflow-mcp
summitflow-mcp provides grant-management calculations for small nonprofits and their AI agents, covering deadline planning, business days, and budgets. Its tools need no API key, account, or network connection.
Key features
- 01Plan key dates backward from grant deadlines
- 02Calculate business days around US federal holidays
- 03Build grant budgets with indirect costs
- 04Value volunteer time
- 05Total in-kind donations at fair market value
- 06Look up US federal holidays
Use cases
- 01Schedule drafting and review dates for a grant application
- 02Count business days during an application period
- 03Prepare a grant budget using a specified cost base
- 04Value volunteer service and donated goods for matching funds
MCP support
As an MCP server, it exposes grant deadline, date, budget, and contribution valuation tools to clients. The tools are pure functions and require no API key, account, or network calls.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
9Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| year | integer | Yes |
Details
From the project READMESummitFlow MCP
Grant management tools for AI agents. Plan backward from a funder deadline, count business days around federal holidays, build a grant budget with indirect on the correct base, and value volunteer and in-kind contributions.
No API key. No account. No network calls. Every tool is a pure function.
Web versions of these tools live at summitflow.co/tools.
Install
Claude Desktop — add to claude_desktop_config.json:
{
"mcpServers": {
"summitflow": {
"command": "uvx",
"args": ["summitflow-mcp"]
}
}
}
Or pip install summitflow-mcp. No environment variables required.
Tools
| Tool | What it does |
|---|---|
grant_deadline_plan | Works backward from a deadline to start-drafting, internal-review and packet-complete dates |
grant_lead_times | The lead times behind that plan, with the reasoning for each |
business_days_between | Business days between two dates, skipping weekends and US federal holidays |
add_business_days | Add or subtract business days from a date |
federal_holidays | US federal holidays for a year, with observed dates |
grant_budget | Full budget with fringe and indirect calculated on MTDC or TDC |
volunteer_time_value | Volunteer hours valued at the published national rate |
in_kind_value | Totals donated goods and services at fair market value |
summitflow_resources | Free templates, guides and calculators on the web |
Why the details matter
Indirect goes on the right base. grant_budget defaults to Modified Total
Direct Cost, which excludes equipment and the portion of each contract above the
first $25,000. Applying your rate to total direct costs instead is the most
common budget error and it inflates the ask in a way reviewers notice. The tool
reports which base it used and what it excluded.
Observed holiday dates, not nominal ones. A federal holiday falling on a Saturday closes offices the Friday before. For a deadline calculation that is the difference between submitting on time and submitting late, so all the date math uses observed dates.
The volunteer rate is cited, not invented. volunteer_time_value defaults
to Independent Sector's published national value of volunteer time (2025), and
says so in its output, because a match figure you cannot source is a figure you
cannot defend. Skilled volunteers doing professional work should be valued at
the rate for that work instead.
Deadlines are not plans. grant_deadline_plan returns three earlier dates
because those are the ones that go on a calendar with a person's name attached.
The deadline itself should be uneventful.
Development
git clone https://github.com/RyanKramer/summitflow-mcp
cd summitflow-mcp
pip install -e .
summitflow-mcp
About
Built by SummitFlow — the grant deadline safety net for small nonprofits. Free templates, guides and calculators.
MIT licensed.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.