Grants Qualification MCP Server
Pricing
from $50.00 / 1,000 tool calls
Grants Qualification MCP Server
Report what a US federal grant agency actually states about who may apply, and screen an applicant type against it. Three outcomes, and 'qualified' is not one of them.
Pricing
from $50.00 / 1,000 tool calls
Rating
0.0
(0)
Developer
NexGen Watch
Maintained by CommunityActor stats
0
Bookmarked
1
Total users
0
Monthly active users
2 days ago
Last modified
Share
Report what a US federal grant agency actually states about who may apply — the coded applicant types, the free-text eligibility narrative, cost-sharing, award range and close date — and screen an applicant type against it.
This server will not tell you that you qualify
It has three screening outcomes and "qualified" is not one of them:
| outcome | meaning |
|---|---|
CLEAR_MISMATCH | the applicant type is not in the agency's coded list — the one negative the coded list supports |
NOT_EXCLUDED_BY_THE_CODED_LIST | the type appears in the list. Not a determination — the narrative may add conditions |
CANNOT_DETERMINE | the coded list defers to the prose, states no types, or none was supplied |
Why. Grants.gov states eligibility in two places and they are not equivalent: a coded
applicantTypes list, and a free-text narrative. The narrative routinely carries geography,
sector, prior-award and partnership conditions no coded list can express — and agencies edit it
roughly 12× more often than the coded list.
The first posted opportunity read during this server's own source check makes the point:
The coded field says, in effect, the rule is not in this field. A server screening on it alone
would answer "eligible applicant type: Others" to an agent asking on behalf of a US nonprofit.
So deferral is detected and reported, and screening returns CANNOT_DETERMINE with the
narrative handed back to read.
Tools
| tool | what it answers |
|---|---|
find_opportunities | Currently posted opportunities, each with its number, close date and Grants.gov URL |
get_eligibility | Everything stated about who may apply, including whether the coded list defers to prose |
screen_applicant | Screen one applicant type — three outcomes, never "qualified" |
explain_outcomes | What each outcome means and why there is no "qualified". Read before acting. |
Scope
US federal discretionary grants on Grants.gov. One jurisdiction. UK Find a Grant was probed and both documented API paths returned 404, so it is not covered and this server does not imply otherwise.
Sources, public and logged-out, no key: api.grants.gov/v1/api/search2 and
api.grants.gov/v1/api/fetchOpportunity. Every outbound call is paced through one shared
strict-pacing bucket at 5 requests/second.
Pricing
| Event | Price |
|---|---|
Server start (apify-actor-start) | $0.05, charged once per server run by the platform |
Tool call (mcp-tool-call) | $0.05 flat per completed tool call |
No other events. Errors and notices are not charged — a bad opportunity id, an empty
applicant type, or a Grants.gov read failure returns without billing. A sourced answer,
including a CANNOT_DETERMINE, is charged: it is the honest answer to the question asked.
If a tool call cannot be billed correctly, the call fails rather than serving free.
