# Andrew's Freebies

> Find evidence-backed U.S. free trials and savings; public benefits are available by explicit scope.

- name: com.andrewsfreebies/mcp
- version: 2.6.0
- connection class: R0 (autonomous)
- trust: 79/100 flags: no-repository
- quality: 85/100 (good)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/com.andrewsfreebies%2Fmcp
- tools: https://api.protogrid.dev/v1/servers/com.andrewsfreebies%2Fmcp/tools

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

## Connection (mcpServers)

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


## Trust

- hygiene: 80
- liveness: 98
- freshness: 85
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-96% ~updated-53d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 85/100 (good)

### protocol: 75 (weight 25)

- warn `protocol.modern`: newest supported version is 2025-06-18; 2026-07-28 not supported, older versions are deprecated until 2027-07-28
- n/a `protocol.stateless`: only applies to 2026-07-28 servers
- pass `protocol.transport`: streamable HTTP
- n/a `protocol.list_ttl`: only applies to 2026-07-28 servers

### 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: 86 (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`: 7 of 7 tools declare readOnlyHint or destructiveHint
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: search (idempotentHint); get_verification (idempotentHint); fetch (idempotentHint); …
- pass `tools.token_cost`: about 1,653 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: 7, about 1,653 tokens to load them all
- tool set hash: 3d8538c6a9fc21d5e55b9ab04aa5e374
- badge: [![protogrid quality](https://protogrid.dev/badge/com.andrewsfreebies/mcp.svg)](https://protogrid.dev/servers/com.andrewsfreebies/mcp)
- 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://andrewsfreebies.com/api/mcp (streamable-http, auth none, reachable true, uptime30d 0.95535713)

## Packages

_none_

## Tools (7)

- `fetch`: Use this after search returns a guide:SLUG, trial:SLUG, or entitlement:SLUG id and the user needs the complete answer, terms, evidence, verification state, or official action path. Do not guess ids, substitute a guide URL for an offer action, or use this for an id from another service.
- `flag_expired_deal`: Use this only when the user supplies an exact existing record id and relevant current first-party evidence showing that its offer ended or is unavailable. Do not use it for rumors, third-party summaries, speculation, or a merely blocked or unreachable page. It writes one non-destructive review signal and never removes or changes the catalog record.
- `get_freshness`: Use this when the user asks about verification freshness for the actionable-candidate corpus, overdue coverage, scheduler liveness, records sharing an audit outcome, or whether one exact non-directory record is still inside cadence. Directory-only leads and public services remain inspectable elsewhere but are excluded from this denominator. For one record, provide both record_type and record_id; its audit outcome still proves only the fields checked. Blocked or unknown never means expired.
- `get_record`: Use this after search_freebies returns an exact trial or entitlement id and the user needs its complete normalized terms, conditions, sources, official action path, or latest verification state. Do not guess ids or treat a directory-only or inconclusive record as confirmed.
- `get_verification`: Use this when the user asks how or when one exact Andrew’s Freebies record was checked, whether its terms matched, or what audit warnings exist. Catalog status and audit outcome are separate; blocked, unknown, and fetch errors never prove expiration, and audits never remove or change records.
- `search`: Use this when the user wants to find, compare, or answer a question about current U.S. free trials, rebates, bonuses, cash back, or other actionable savings. It returns citable answer guides and normalized records. Fetch the best returned id before giving detailed terms. Do not use it for purchases, account access, or guaranteed eligibility.
- `search_freebies`: Use this when the user needs explicit corpus scope, result limits, or a rich normalized list rather than the standard citable search/fetch flow. It is read-only: the default uses site-controlled actionable U.S. trials and savings, while only an explicit open-source scope reads public GitHub. Use public-benefits only for eligibility-based services. Do not use it for purchases, account access, or guaranteed eligibility.