# Overture Maps docs: AI Agent Readiness Score 58.8% (F) **58.8% · 47/80 · AI Agent Readiness Score · 27/30 reading points · 20/50 agent surface points** Overture Maps received 12 PASS votes and passed 2 of five agent surface checks. The clearest finding came from the find the exact access limits task. - Tested: 2026-08-20 - Published: 2026-09-21 - Battery: v1 - Scoring: reading 30 pts · surface 50 pts - Docs: https://docs.overturemaps.org/getting-data/ Three AI models, GPT 5.6 Sol, Claude Opus 5, and DeepSeek v4 Flash, each read Overture Maps’s public documentation independently and attempted five first-hour developer jobs: download the first map extract, find the exact access limits, recover from a release data change, authenticate AWS catalog access, use the Python client. No accounts, API calls, or code execution were used. Every verdict came from public pages and every published quotation passed a live verification check. Preflight substitutions: T3: Monthly changelog diagnosis replaces HTTP 429. T4: AWS Athena authentication replaces webhooks. ## Freshness [How rechecks work](https://docsforagents.com/methodology/#freshness) - Category: [Geospatial & maps](https://docsforagents.com/grades/?category=geospatial) - Tested: 2026-08-20 - Quotes verified: 2026-08-20 - Surface rechecked: 2026-09-21 No change since the test. 5 of 5 quoted passages still appear on the live pages. ## Agent surface checks · 20/50 | Check | Verdict | Points | | --- | --- | --- | | llms.txt | PASS | 10 | | llms-full.txt | PASS | 10 | | Markdown mirror | FAIL | 0 | | MCP server | FAIL | 0 | | Docs AI | FAIL | 0 | ## The Reading Test · 27/30 | Task | GPT 5.6 Sol | Opus 5 | DeepSeek v4F | Consensus | | --- | --- | --- | --- | --- | | Download the first map extract | PASS | PASS | PASS | PASS | | Find the exact access limits | PASS | PARTIAL | PASS | PASS | | Recover from a release data change | PARTIAL | PASS | PASS | PASS | | Authenticate AWS catalog access | PASS | PASS | PASS | PASS | | Use the Python client | PASS | PARTIAL | PASS | PASS | Docs platform: Docusaurus 3.10.2 (unscored) · verified 2026-08-20 ## What to fix first These 3 fixes could add up to 30 points to the AI Agent Readiness Score. The list ranks each fix by the points it would add. [How the ranking works](https://docsforagents.com/methodology/#what-to-fix-first) 1. **+10 points · Markdown mirror · check failed** **Found:** Tested Markdown and content-negotiation routes returned HTML or HTTP 404. **Fix:** Serve every docs page as Markdown at its URL plus .md, and return the same Markdown for Accept: text/markdown. **Evidence:** [docs.overturemaps.org/getting-data/cloud-sources.md](https://docs.overturemaps.org/getting-data/cloud-sources.md) 2. **+10 points · MCP server · check failed** **Found:** Initialize returned an HTML error, not an MCP response. **Fix:** Run a public MCP server for the docs that answers a JSON-RPC initialize call and offers a docs search tool. **Evidence:** [docs.overturemaps.org/mcp](https://docs.overturemaps.org/mcp) 3. **+10 points · Docs AI · check failed** **Found:** The live docs expose standard search but no public assistant control. **Fix:** Add an assistant to the docs site that answers questions from the docs and links to its sources. ## What the docs get right - **Download the first map extract: 3 PASS votes.** The official documentation source gives one complete bounded building download in GeoJSON format. - **Authenticate AWS catalog access: 3 PASS votes.** The Athena guide identifies the cross-account setup path and exact Overture catalog ID. - **2 of 5 agent surface checks.** Present: llms.txt, llms-full.txt. Missing: markdown mirrors, an MCP server, docs AI. ## Download the first map extract **PASS** PASS consensus from 3 PASS. The page gives pip installation and one complete GeoJSON building command with a four-coordinate bounding box and output file. The quickstart and the Python Client page give the same install-plus-download sequence with a complete building bounding-box example, so no step requires guessing. Installation via pip (pip install overturemaps) and the full download command with bounding box, format, type, and output path are documented on the same page with no ambiguity. ## Find the exact access limits **PASS** PASS consensus from 2 PASS, 1 PARTIAL. The client page fixes order, formats, and whole-type behavior; the cloud-sources table lists six themes and 15 types. Bounding-box order, the three output formats, and whole-type behavior are exact, but the complete list of six themes and fourteen feature types lives on a different page than the CLI reference, and the CLI's own accepted --type values are deferred to a runtime command with "Run overturemaps download --help for the complete list." All six themes and their 15 feature types are tabulated on the cloud-sources page; output formats (geojson, geojsonseq, geoparquet), bounding-box order (west, south, east, north), and whole-type behavior (omit bbox to download the entire type) are documented precisely with no contradictions. ## Recover from a release data change **PASS** PASS consensus from 2 PASS, 1 PARTIAL. The docs explain changelog queries and classes, but the latest release notes disclose an incorrect added classification. The changelog page defines all four change_type values, gives the current changelog path, and supplies runnable DuckDB and Athena queries that join changelog rows to release features by id, though its worked examples are still pinned to the 2024-06-13-beta.0 release. The changelog page documents change types, provides DuckDB queries to detect changes, shows how to join with release data via id, and confirms changelogs persist beyond the 60-day data retention window, enabling full diagnosis. ## Authenticate AWS catalog access **PASS** PASS consensus from 3 PASS. The Overture and AWS pages provide the region, IAM policy, catalog registration steps, source name, and release SQL. The page supplies the region, the cross-account Glue catalog ID, a complete copy-pasteable IAM policy, the release table naming convention, and a working query, but it does not mention the first-run Athena query result location that AWS requires before any query executes. The Athena example page provides the exact cross-account catalog ID, a complete IAM policy JSON with glue:GetTable*/GetDatabase*/GetPartition* actions, the required region (us-west-2), and query examples for release tables. ## Use the Python client **PASS** PASS consensus from 2 PASS, 1 PARTIAL. The linked official README gives minimum library calls, matching writer formats, and the same latest-release STAC default. The page names the library, documents install paths, output formats, and default STAC behavior via the --stac/--no-stac flag, but every example on it is a shell command, so a minimum in-process library call cannot be written from the docs and must be recovered from the linked overturemaps-py repository. The overturemaps library is listed on PyPI and documented on the docs site; both pages agree on installation, CLI usage, the three supported output formats (geojson, geojsonseq, geoparquet), bounding-box order, and default STAC catalog behavior. ## The receipt > West, south, east, north coordinates. Omit to download the entire type. The client page fixes order, formats, and whole-type behavior; the cloud-sources table lists six themes and 15 types. - [docs.overturemaps.org/getting-data/overturemaps-py/](https://docs.overturemaps.org/getting-data/overturemaps-py/) ## Agent surface notes Initialize returned an HTML error, not an MCP response. The live docs expose standard search but no public assistant control. ## Method note This is a reading test of public documentation, not an execution test. No accounts were created and no API calls were run. The AI Agent Readiness Score counts fifteen reading votes at PASS 2, PARTIAL 1, and FAIL 0, for 30 possible points. Five agent surface checks add 10 points each. The total is 80. Consensus chips show each row majority and do not affect scoring. The panel split on 3 of five tasks. Quotes shown here were re-fetched and confirmed verbatim on 2026-08-20. Methodology: https://docsforagents.com/methodology/ Canonical URL: https://docsforagents.com/reports/overture-maps-docs-ai-agent-readiness/