kaiserpermanente.org
Gemini · gemini-2.5-flash · HTTP · Target kaiserpermanente.orgA measurement at one moment, not a verdict · MethodologyAI agents can access the site reliably but fail to complete most tasks fully due to missing machine-readable navigation files.
Publish llms.txt+777 → 84
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 kaiserpermanente.org without encountering bot blocks, achieving a 95% reliability score and citing the site's own URLs 100% of the time. However, the overall task success rate was only 20%, with 8 out of 10 intents resulting in partial success. This indicates that while agents can reach the site, they struggle to find complete answers for critical tasks like booking, finding locations, and checking fees. The complete absence of agent-facing navigation aids like `llms.txt` and `sitemap.xml` forces agents to rely on raw HTML crawling, which is inefficient and leads to a very low link following score of 12%. Providing these standard files would significantly improve the agents' ability to navigate the site and extract complete information.
What works
- The site does not block AI agents, allowing native access to content.
- Agents consistently cite the site's own URLs (100% own source rate).
- High on-site discovery (95%) and reliability (95%) scores indicate stable infrastructure.
- Full success was achieved for eligibility and careers 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
- robots.txt contains Disallow rules
- Cookie consent banner: OneTrust
- First byte after 337ms
Evidence
- /llms.txt — absent (HTTP 404)
- /llms-full.txt — absent (HTTP 404)
- /sitemap.xml — absent (HTTP 404)
- /sitemap-index.xml — absent (HTTP 404)
- /sitemap_index.xml — absent (HTTP 404)
- /robots.txt — present
Evidence
- 102 links in the raw HTML
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 100%
- 2xx rate 100%
- done_unaided
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 100%
- 2xx rate 100%
- done_unaided
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 100%
- 2xx rate 100%
- done_unaided
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 80%
- 2xx rate 80%
- done_unaided
Evidence
- outcome success · verdict full
- the final answer cited this site
- site-guided navigation 100%
- 2xx rate 100%
- done_unaided
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 80%
- 2xx rate 90%
- done_unaided
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 100%
- 2xx rate 100%
- done_unaided
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 75%
- 2xx rate 75%
- done_unaided
Evidence
- outcome success · verdict partial
- the final answer cited this site
- site-guided navigation 100%
- 2xx rate 100%
- done_unaided
Evidence
- outcome success · verdict full
- the final answer cited this site
- site-guided navigation 50%
- 2xx rate 100%
- done_unaided
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 | |
|---|---|---|---|---|---|---|
| book an appointment | Partial | 100% | 100% | 100% | Yes | |
| find a branch | Partial | 100% | 100% | 100% | Yes | |
| fees & rates | Partial | 100% | 100% | 100% | Yes | |
| get support | Partial | 100% | 80% | 80% | Yes | |
| am I eligible | Full | 100% | 100% | 100% | Yes | |
| contact a human | Partial | 100% | 90% | 80% | Yes | |
| refund policy | Partial | 100% | 100% | 100% | Yes | |
| find a document | Partial | 100% | 75% | 75% | Yes | |
| latest notice | Partial | 100% | 100% | 100% | Yes | |
| open roles | Full | 50% | 100% | 50% | Yes |
About this report
- Measured
- 9/10/2026, 10:09:16 PM 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.