Radar

# Fix list: Radar

From Anchor Terminal's listing at https://www.anchorterminal.com/tools/radar, the October 2026 research run, assessed 9 October 2026. Grade D, 53.3 out of 100.

This is everything the published grade says the listing lacks, the biggest possible gain to the total first. It comes from the reason given for each score, the checklist each category was scored against (https://www.anchorterminal.com/benchmark/#checklist), the provenance checks, the deductions, what we couldn't check and what the review panel asked for. A fix counts at the next check, once it's public.

For a coding agent working on Radar: work through the items below in the product, its docs and its public pages. Each category gives the reason for its score, with the points each checklist item earned, and the checklist itself, so the gap is the items that earned less than their points. Change the product, not the wording, and keep a note of what you changed and where it's published.

## 1. Payments & pricing, 10 out of 100, up to 11.3 more on the total

Why it scored 10: No x402, MPP or L402 (0). The pricing page publishes starting prices without a login, $0.50 per 1,000 calls for Core Maps and $2.00 for Premium Maps, but every plan is an annual commitment at a rate set by quote (10). The pricing page says Radar has no free tier, and the technical validation period goes through sales (0). Keys come from a dashboard account, and the sign-up page was closed to us by robots.txt, so no programmatic route was found (0).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-payments):

The published rubric, also on the [x402 page](https://www.anchorterminal.com/x402/).

- 40, a machine payment protocol (x402, MPP or L402) on the tool's own endpoints. 10 to 30 when it covers only some endpoints or only goes through a third party, and the note says which.
- 20, per-call or per-unit pricing published without a login. 10 for public plan-only pricing, 0 for "contact sales" or prices behind a login.
- 20, a free tier or trial that doesn't need a card.
- 20, autonomous onboarding, meaning an agent can get access without a person signing up in a browser (keyless use, x402, a programmatic key API).

Payment platforms and agent wallets rarely charge for their own API over a machine protocol, so the first line has steps for them, and the highest one that applies counts. 40 when x402, MPP or L402 runs on all their own endpoints, 30 when it runs on part of their own API, 25 when their merchants can accept one, 20 for running a facilitator, 15 for paying as a buyer, and 0 when the only protocol is their own. Merchant acceptance sits above a facilitator because the platform's own customers can charge agents through it, while a facilitator settles for sellers who wire up the protocol themselves. The counter-argument (a facilitator does more for the protocol as a whole) has a point. Each note says which step applied.

Open-source software you run yourself is scored on its hosted or paid option if it has one. A free, self-hosted package with nothing to buy gets 20, 20 and 20 for the last three lines, and 0 to 40 for the first only if it ships a payment protocol.

## 2. Schema & documentation, 50 out of 100, up to 8.1 more on the total

Why it scored 50: No OpenAPI description or other machine-readable contract was found in the docs index (0). docs.radar.com publishes llms.txt and a Markdown twin of each page, with the whole API reference in one 148 KB file (10). Descriptions say what an operation is for and sometimes when to use another, for example forward geocoding for complete addresses and autocomplete for partial ones (14). Parameters are listed with type, required or optional, ranges and allowed values, though coordinate lists travel as pipe-delimited strings and there is no schema to validate against (11). Every operation has a sample request and response, and there is one list of 12 response codes with one sample error (11). Paths carry `/v1`, and no API changelog was found (4).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-schema):

APIs and MCP servers.

- 25, a machine-readable contract (a public OpenAPI file or similar; for MCP, typed JSON Schema inputs on every tool).
- 10, llms.txt or Markdown docs served for agents.
- 0 to 20, descriptions that say what a tool is for, when to use it and when not to, read from the tool definitions in the source or the API reference.
- 0 to 15, typed inputs with enums, constraints and required fields, and no free-form JSON blobs.
- 0 to 15, examples and documented error responses.
- 15, versioning and a public changelog.

Models are read from the API reference, the OpenAPI file, llms.txt, the structured-output and tool-use docs and the model cards. Frameworks from docs a model can follow, typed interfaces, examples and the API reference.

## 3. Security & auth, 56 out of 100, up to 7.7 more on the total

Why it scored 56: Four keys a project, test and live, publishable and secret, sent in the `Authorization` header only. A secret key can read or write any data and no scopes were found. Key rotation was not read (20). Publishable keys reach only 16 client-safe operations and can be restricted by website domain or mobile app identifier, and dashboard accounts have read, write, admin and owner roles. No confirmation step for deletes (10). Results are addresses and place names from open and commercial datasets, with no prompt-injection guidance (6). Audit logs cover dashboard requests on enterprise plans, and each API response has a request ID (7). security.txt answers 404. The FAQ gives a disclosure address, and the DPA states SOC 2 Type II, a yearly third-party penetration test and a bug bounty, with the report on request. The trust centre is script-drawn and was not read (13).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-security):

- 0 to 30, the credential model. 30 for OAuth 2.1 with scopes, or scoped and revocable keys with rotation. 20 for plain revocable API keys. 10 for one all-powerful key. 10 off when a secret can travel in a URL query string as a documented option.
- 0 to 20, read-only or least-privilege modes, and confirmation or approval for destructive actions.
- 0 to 15, prompt-injection posture where the tool returns untrusted content (documented mitigations or guidance). A tool that returns no untrusted content gets 10.
- 0 to 15, audit logs or per-call visibility for the operator.
- 0 to 20, a security programme. security.txt or a disclosure policy, a bug bounty, SOC 2 or ISO 27001, advisories handled in public.

Models are read for retention, whether API data trains models (and whether that's off by default), zero-retention options and certifications. Frameworks for telemetry defaults, approval hooks, guardrails and sandboxing.

## 4. Reliability, 70 out of 100, up to 6 more on the total

Why it scored 70: Read with the hosted lines. status.radar.com is a Statuspage site with components for the Geocoding, Search, Routing and Maps APIs (20). Its history feed lists two incidents in the 90 days to 9 October 2026. Elevated API errors and latencies for 3 minutes on 14 July, and elevated API errors on 27 August from 19:31 to 20:34 EDT, with the share of failed requests not stated, so we read both as minor (20). Each operation states a default rate limit, 100 requests a second for geocoding down to 10 for matrix (15). 429 is documented with three meanings, with no Retry-After header or backoff guidance found, and PUT upserts keyed on `tag` and `externalId` make geofence writes repeatable (5). No SLA is published (0). The APIs carry no beta label (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-reliability):

Hosted APIs, MCP servers, models and platforms.

- 20, a public status page with component history (Statuspage, Instatus, BetterStack or the vendor's own).
- 0 to 30, the incident record for the last 90 days on that page. 30 for a clean record or trivial incidents only, 20 for minor incidents only, 10 for one major outage (an hour or more of a core API down, or errors across the board), 0 for several. 5 when there's no history we could read, and the note says so.
- 15, rate limits documented with numbers.
- 15, documented 429 or overload handling (Retry-After, backoff guidance), and idempotency keys or safe-retry guidance where writes are involved.
- 10, an SLA published for any paid tier.
- 10, the surface agents use is generally available, not beta or preview.

Local packages, SDKs, frameworks and stdio MCP servers.

- 20, installs from an official package with supported runtimes stated.
- 25, a public CI and test suite, passing on the default branch.
- 0 to 25, open crash or regression issues relative to activity (25 for few and handled, 0 for many, old and unanswered).
- 15, semver discipline and breaking changes called out in a changelog.
- 15, version 1.0 or later, or declared stable.

Protocols are read from their reference implementations, the public facilitators or servers, spec stability and test vectors.

## 5. Agent ergonomics, 65 out of 100, up to 5.7 more on the total

Why it scored 65: Read as an API. `limit` on search operations (1 to 100), `layers` and `country` filters on geocoding, and optional `geometry` on route optimisation size a response, with no field selection (14). List operations page with `limit` up to 1,000 and date cursors such as `updatedBefore` (16). Errors are JSON with `code`, `param` and `message`, and every response carries `X-Radar-Request-ID` (15). The maps operations only read. Geofence and beacon writes are PUT upserts by external ID, and no idempotency key was found for trip or user creation (10). Most maps calls need one parameter. Official SDKs cover web, iOS, Android and five cross-platform frameworks, all client-side, with no server library found (10).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-ergonomics):

- 0 to 25, context cost. For MCP, the number and size of the tool definitions (25 for ten or fewer compact tools, 15 for 11 to 30, 5 for more than 30, plus up to 10 back for toolsets, dynamic loading or read-only subsets). For APIs, whether responses can be sized (field selection, limits, summaries).
- 20, pagination, filtering and output-size controls.
- 20, actionable, documented error responses, codes and messages an agent can recover from.
- 20, idempotency or safe retries, and for MCP the `readOnlyHint` and `destructiveHint` annotations.
- 15, sensible defaults, few required parameters, and official SDKs in at least two languages.

Models are read for tool use, structured output, prompt caching, context length, batch and SDKs. Frameworks for how much code and how many defaults a tool-calling agent with MCP needs.

## 6. Transparency & trust, 55 out of 100, up to 3.9 more on the total

Made of editorial 51, provenance 58.

Why it scored 55: Closed service with public Terms of Use, and an Apache-2.0 web SDK. The terms date from 24 May 2023, describe a free version the pricing page says does not exist, and paid plans run on a quoted agreement that is not published (12). The privacy notice of 13 May 2025, a public DPA and the FAQ agree on a default retention of 1 year, with 90 days for trips and 7 days for location updates in the FAQ. The privacy notice itself gives no period (22). No deprecation policy was found, and the terms let Radar change or stop the services without notice. One parameter is marked deprecated with no date (2). The DPA names AWS and MongoDB as sub-processors, and the privacy notice says hosting is in the United States (15).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-transparency):

- 0 to 30, source availability and licence clarity. 30 for open source under an OSI licence, 15 for closed with clear terms, 0 for unclear terms.
- 0 to 30, data handling and retention statements that agree with each other (privacy policy, DPA, retention periods, subprocessors).
- 0 to 20, a deprecation policy or notices with dates.
- 0 to 20, telemetry disclosed with an opt-out (local software), or subprocessors and data locations disclosed (hosted).

The other half of Transparency and trust is the provenance score, computed from checked facts (below). The category score is the mean of the two.

Provenance checks not met in full (half of this category, computed from checked facts):

- Endpoint on the vendor's domain: api.radar.io is not on radar.com (0 of 15)
- Terms of service: read, states 4 of the 7 things a reader expects, and has 2 clauses that cost points (3.4 of 10)
- Changelog: not found (0 of 10)
- security.txt: not found (0 of 10)

## 7. Maintenance & community, 77 out of 100, up to 2 more on the total

Why it scored 77: No API changelog was found, so recency rests on Radar's own SDK. radarlabs/radar-sdk-js tagged autocomplete-v1.1.1 on 14 September 2026 and a 5.3.0 beta on 29 September, within 30 days (30). Stable tags in the last 90 days are v5.1.2 on 25 August, v5.1.3 on 31 August, v5.2.0 on 1 September and autocomplete-v1.1.1 (20). No public changelog for the API, support through a customer success manager and a feedback form, and GitHub issue replies were not read (5). Official SDKs for eight platforms are current (15). The web SDK repository carries Jest tests and CodeQL and npm publish workflows, whose results were not read (7).

The checklist (https://www.anchorterminal.com/benchmark/#checklist-maintenance):

- 0 to 30, time since the last release, or the last published model or API change for a closed service. 30 within 30 days, 20 within 90, 10 within 180, 0 older.
- 20, at least three releases or dated changelog entries in the last 90 days.
- 0 to 25, responsiveness. Issues and pull requests answered on GitHub (the open issues and how recent the replies are). For closed services, a public changelog and a support or community channel that answers, 0 to 15.
- 15, presence in the official MCP registry under a verified namespace (MCP servers), or current official SDKs (APIs and models).
- 10, package health, current dependencies and CI.

Models are read for deprecation notice periods and model churn rather than release counts.

## Deductions

Each comes off the total. A fixed and documented problem counts for less at the next check.

- 9 October 2026. The Maps product page says the geocoding APIs come with 'the right to store, reuse, and export results freely' (https://radar.com/product/maps), while section 2(d) of the Terms of Use forbids storing address or point-of-interest data from the geocoding, autocomplete or place search APIs for more than 30 days (https://radar.com/terms) and the geocoding docs repeat the 30-day limit (https://docs.radar.com/maps/geocoding.md). 2 points for a claim the published terms contradict. A quoted enterprise agreement may differ and is not published.

## What we couldn't check

What we couldn't read counted as absent. Publishing it on a page a plain HTTP fetch can read (not only in a browser) lets the next check count it.

- unchecked: the trust centre at trust.radar.com, which is script-drawn
- unchecked: whether sign-up at dashboard.radar.com is self-serve, and key rotation, because that host's robots.txt disallows every path
- unchecked: the per-endpoint billing page the pricing FAQ links, the search, routing and maps docs pages, and the tutorials
- unchecked: GitHub stars, npm downloads, issue response times and CI results for radar-sdk-js
- unchecked: who registered radar.io, the domain the API answers on. Nothing was sent to the API host
- Whether a quoted enterprise agreement lifts the 30-day storage limit in the public terms, as the Maps product page implies
- Whether the free version the 2023 terms describe is still open to new accounts. The pricing page says there is no free tier
- The share of requests that failed in the incident of 27 August 2026 is not stated on the status page
- The terms forbid using the services to develop or test other geofencing, geocoding, autocomplete or place search APIs. This matters before any probe is run
- The lead was right on the endpoints and key types. It did not say that the API host is api.radar.io or that access is sold by quote with no free tier

## Weaknesses

- The pricing page says Radar has no free tier and every plan is an annual commitment set by quote
- No OpenAPI description and no API changelog were found in the docs index
- The terms cap storage of geocoding, autocomplete and place results at 30 days, while the Maps product page claims a right to store results freely
- The terms of 24 May 2023 let Radar change or stop the services without notice, and still describe a free version
- A secret key can read or write any project data, with no scopes found
- The terms forbid using the services to develop or test other geocoding, autocomplete or place search APIs, which matters before any probe is run

## What costs an agent a turn today

The notes we give agents before they call it. Each one is a workaround an agent shouldn't need.

- Send the key in the `Authorization` header with no `Bearer` prefix, and call `https://api.radar.io/v1`, not radar.com
- Use a publishable key for geocoding, search and routing. Keep the secret key for users, trips, geofences and beacons
- Call `/v1/search/autocomplete` for partial addresses and place names. `/v1/geocode/forward` is documented as best for complete addresses
- Set `units=metric` on routing calls. The default is imperial, with distances in feet
- Treat 429 as one of three things (rate limit, no state change or selective throttling) and 402 as usage exceeded. No Retry-After header is documented
- Do not keep geocoding, autocomplete or place results for more than 30 days

## When it's done

Send what changed and where it's published as a dispute (https://www.anchorterminal.com/builders/#disputes, or `POST https://www.anchorterminal.com/api/v1/contact` with `"kind": "dispute"`). Disputes are answered in public, and the listing is checked again by the same checklist. Paying for an audit or a listing claim changes nothing here.