# MissedTakes

> Graded track records of NFL and election pundits: records, leaderboards, takes, game and race calls.

- name: co.missedtakes/missedtakes
- version: 1.0.0
- connection class: R0 (autonomous)
- trust: 83/100 flags: no-repository
- quality: 96/100 (strong)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/co.missedtakes%2Fmissedtakes
- tools: https://api.protogrid.dev/v1/servers/co.missedtakes%2Fmissedtakes/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

```json
{
  "mcpServers": {
    "missedtakes": {
      "type": "http",
      "url": "https://missedtakes.co/api/mcp"
    }
  }
}
```


## Trust

- hygiene: 80
- liveness: 100
- freshness: 100
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-100% +updated-10d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 96/100 (strong)

### protocol: 88 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- warn `protocol.list_ttl`: tools/list declares ttlMs 0, so clients re-fetch it on every use

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 100 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- pass `tools.annotations`: 9 of 9 tools declare readOnlyHint or destructiveHint
- pass `tools.directory_hints`: every tool declares readOnlyHint, destructiveHint, idempotentHint and openWorldHint
- pass `tools.token_cost`: about 2,147 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: 100 (weight 15)

- pass `stability.changes`: no tool changes in 30 days
- pass `stability.rug_pull`: no tool changed its meaning under the same name

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 9, about 2,147 tokens to load them all
- tool set hash: a01b0dc11207b7a3e8f1d7f3e1f2a787
- badge: [![protogrid quality](https://protogrid.dev/badge/co.missedtakes/missedtakes.svg)](https://protogrid.dev/servers/co.missedtakes/missedtakes)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

_No tool definition changes recorded._

## Remotes

- https://missedtakes.co/api/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (9)

- `election_pundit`: One pundit's election record: their graded record from past cycles (accuracy, MissedTakes Score, sample size) and their calls in the current cycle, newest first. Look up by slug, or by part of their name.
- `election_race_calls`: The individual calls on one race or chamber, newest first: who said it, when, what they picked or the number they named, a short excerpt of their wording, and the pundit's graded election record from past cycles. raceSlug comes from election_races, or is chamber:senate, chamber:house or national:house-vote.
- `election_races`: The races, chamber control and the national House popular vote that pundits have called in an election cycle (the current one unless you pass cycle), most-called first: how many pundits called each, how the field splits (each pundit's latest pick counted once, and split into forecasters and commentators), calls in the last 7 days, and the VoteHub polling average where one exists. Filter by office (president, senate, house, governor, chamber, national) or two-letter state. Use it to find a race slug.
- `find_entities`: Look up the slug MissedTakes uses for a pundit, a publication (outlet), an NFL team, an NFL player or a prediction type, by full or partial name. Call this first whenever the user names someone: the other tools take slugs, never names. Only pundits and outlets with a public page are returned.
- `game_room`: What the pundits predicted for one NFL game, counted the way its page on missedtakes.co counts it: how many took each side (one voice per pundit, their latest call standing), the median predicted margin, each pundit's latest score call as a home-minus-away margin, and the final score once the game is played. Identify the game by season year, week (1-18 regular season, 19-22 playoffs) and either team's slug.
- `leaderboard`: The ranked NFL pundit leaderboard, the same rows missedtakes.co/nfl/leaderboard prints: MissedTakes Score (skill against a naive baseline, which is the ranking), accuracy, graded takes and articles. Provisional records (under ten graded articles) are not ranked. Pass a prediction type slug (find_entities, kind prediction_type) to rank within one kind of claim, such as mock drafts.
- `list_takes`: Individual NFL predictions: who made them, when (claimedAt is the publication date), a short excerpt of the wording, the grade (accuracy 0-100, or pending) and the grading arithmetic. Filter by pundit, outlet, prediction type, team, player, season, verdict (hit is 80+, close is 40-79, miss is under 40) and claim dates. 25 rows a page; pass page for more.
- `outlet_record`: One publication's NFL record, aggregated across everyone writing or speaking under its banner: overall and by prediction type, with how many pundits contribute.
- `pundit_record`: One pundit's published NFL record: overall MissedTakes Score, accuracy and sample size, the record broken down by prediction type, and any Pundit of the Week or Month awards. For their election record, call election_pundit.