uber.com
Gemini · gemini-2.5-flash · HTTP · Target uber.comA measurement at one moment, not a verdict · MethodologyAI agents can access Uber.com and consistently cite its own sources, but aggressive rate limiting and missing agent-specific files force them to rely on external search, resulting in partial failures for key tasks.
Publish llms.txt+774 → 81
All of them together: 95 (A)
Projected from the audit record. After fixing, re-run the task or re-measure to confirm.
Agents successfully accessed the homepage and consistently cited Uber's own URLs across all tasks, achieving a 100% own-source rate. However, the overall task success rate was only 40%, with six out of ten intents resulting in partial success. The site returned HTTP 429 (Too Many Requests) errors for standard discovery files like sitemap.xml and robots.txt, indicating bot protection mechanisms are hindering automated exploration. Furthermore, the absence of an llms.txt file meant agents lacked direct guidance, forcing them to guess URLs or rely on external search bridges for tasks like getting started, contacting support, and comparing options. Consequently, agents encountered errors and failed to retrieve complete information for critical user intents such as pricing, policies, and locations.
What works
- The homepage is accessible to agents (HTTP 200) and provides a healthy number of internal links (135 anchors).
- Agents consistently cited the site's own URLs (100% own source rate) across all tested intents.
- Full success was achieved for signup, API documentation, getting started, and contact intents.
What to fix
For each item: its current state, why it matters, how to fix it and how to check. The same content comes as files you can hand straight to whoever does the work.
Putting the items together…
Fix files
Files generated from this audit. Paste them in, or hand them to a coding agent. Have someone review them before they go live.
Building the fix files…
Detailed results
The results behind the score, and the measurement record.
Evidence
- homepage HTTP 200
- redirect hops: 2
Evidence
- no bot blocking observed
- /robots.txt absent
- First byte after 230ms
Evidence
- /llms.txt — absent (HTTP 404)
- /llms-full.txt — absent (HTTP 429)
- /sitemap.xml — absent (HTTP 429)
- /sitemap-index.xml — absent (HTTP 429)
- /sitemap_index.xml — absent (HTTP 429)
- /robots.txt — absent (HTTP 429)
Evidence
- 135 links in the raw HTML
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 33%
- 2xx rate 67%
- done_big_detour
Evidence
- outcome success · verdict full
- the final answer cited this site
- site-guided navigation 33%
- 2xx rate 100%
- done_big_detour
Evidence
- outcome success · verdict full
- the final answer cited this site
- site-guided navigation 67%
- 2xx rate 100%
- done_big_detour
Evidence
- outcome success · verdict full
- the final answer cited this site
- site-guided navigation 0%
- 2xx rate 100%
- done_big_detour
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 33%
- 2xx rate 67%
- done_big_detour
Evidence
- outcome success · verdict full
- the final answer cited this site
- site-guided navigation 0%
- 2xx rate 50%
- done_big_detour
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 50%
- 2xx rate 100%
- done_big_detour
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 67%
- 2xx rate 100%
- done_big_detour
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 50%
- 2xx rate 100%
- done_unaided
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 0%
- 2xx rate 100%
- done_big_detour
WebMCP — what the site declares about itself
This audit ran with the WebMCP check switched off. That does not mean there are no tools — it means we did not look. Turn on "Check WebMCP" when re-auditing.
Act-stage check (nothing submitted)
Checks whether an agent could fill in the forms and press the buttons, up to the point of pressing. No form was submitted, nothing added to a cart, nothing paid. Report only; not scored.
Per-task results
| Task | Verdict | Own-site share | Request success | Followed guidance | Cited own site | |
|---|---|---|---|---|---|---|
| find pricing | Partial | 100% | 67% | 33% | Yes | |
| create an account | Full | 100% | 100% | 33% | Yes | |
| find api | Full | 100% | 100% | 67% | Yes | |
| get started | Full | 100% | 100% | 0% | Yes | |
| get support | Partial | 100% | 67% | 33% | Yes | |
| contact a human | Full | 100% | 50% | 0% | Yes | |
| find a branch | Partial | 100% | 100% | 50% | Yes | |
| refund policy | Partial | 100% | 100% | 67% | Yes | |
| latest notice | Partial | 100% | 100% | 50% | Yes | |
| compare two options | Partial | 100% | 100% | 0% | Yes |
About this report
- Measured
- 9/11/2026, 3:16:07 AM GMT+9
- Engine
- Gemini · gemini-2.5-flash · HTTP
This is what AI agents could do on the site at the time above, measured by actually using it. It is not a verdict on the site or the company, and the result can change when the site does.