Your agent alone, or your agent with OpenAdapt
Give your agent a program, not a mouse. OpenAdapt compiles your agent's run of a task into a program the agent can call for later runs. The program repeats the same steps without a model and reads each save back. In timed referral runs, it saved 62 of 63 correctly, as the agent alone did, and had no model fees.
Updated 2026-11-19
What OpenAdapt adds to your agent
| Your agent alone | Your agent with OpenAdapt | |
|---|---|---|
| First run of a task | The agent drives the screen | The agent drives the screen. OpenAdapt records the run, reads the save back from the system of record, and compiles it |
| Later runs | Model calls, screenshots to a provider, a different path each time | The compiled program repeats the same steps without model calls or screenshots, and reads each save back |
| Before each save | The model judges from the screen | From the compiled program on, the right record is confirmed first |
| When the app changes | The agent improvises on live data | The run stops before saving; your agent shows the fix and a person approves it |
| When a save may not have landed | The agent judges from the screen, and may retry | Your agent gets "not sure if saved", and nothing is retried blindly |
| What comes back | A transcript | An outcome your agent can act on, and a signed receipt when the save was done and checked |
OpenAdapt adds little to a task that you'll run once and that saves nothing.
What your agent brings
- It can start a new or loosely specified task from a plain-language goal, without anyone recording it first.
- It can work through screens it has never seen, and choose its next step from what's in front of it.
- It improves with each model generation, without changing your task.
- On simple tasks the agent alone can be as correct as a program. On the public OpenEMR demo the agent passed 10 of 10 runs, and OpenAdapt's program passed 19 of 20.
OpenAdapt keeps all of that for the first run and for repairs, and runs the repeats itself, so the agent doesn't have to drive the screen again.
What we measured
| Test | Agent alone | OpenAdapt program | How it was judged |
|---|---|---|---|
| openIMIS claim entry | Claude Sonnet 5 (claude-sonnet-5), 21 Jul 2026: 3 of 3 correct; $0.48 a run; 69.8 s mean | July 2026: 3 of 3 checked; 0 model calls; 25.6 to 30.3 s | A database read. Different days, not the same baseline |
| OpenEMR public demo, add a note | Claude Sonnet 5 (claude-sonnet-5), 8 Jul 2026: 10 of 10; about $0.55 a run; 70.4 s median | 19 of 20; 0 model calls; 39.2 s median | The same screen check of the saved note for both |
| Frappe Lending loan application | Claude Sonnet 5 (claude-sonnet-5), 21 Jul 2026: 6 of 6 correct writes, 5 of 6 clean; $0.42 a run | 16 Jul 2026: 6 of 6 checked | Database reads. Not the same baseline |
| Local OpenEMR new patient | Claude Sonnet 5 (claude-sonnet-5), 21 Jul 2026: 0 of 6; each trial used up its action budget before saving; $0.89 a run | Not run in that test | A database read |
| Local OpenEMR referral entry, timed (method) | Claude Sonnet 5.5 (claude-sonnet-5-5), 14 to 15 Oct 2026: 62 of 63 saved and matched; 36.5 s median; about $0.08 a run (median); no new chart on the possible duplicate in 9 of 9 | 62 of 63 saved and matched; 0 model calls; 29.7 s median; stopped on the possible duplicate in 9 of 9 | The same database read-back by our test setup for both; medians over the 62 correct runs each |
How we timed the referral runs
We used a local OpenEMR 8.0.0.3 test copy with 8 synthetic referrals, one of which was a possible duplicate of an existing patient. OpenAdapt replayed a program learned from a demonstration that finds each field by the page's structure (0 model calls). The agent was Claude Sonnet 5.5 (claude-sonnet-5-5) in our capped harness, at list-price fees. Runs were interleaved in blocks, 14 Oct 22:27 to 15 Oct 05:35 ET, with the browser already open (warm). The clock covers sign-in, the entry and the read-back. Each result was checked by reading the new chart from the database. You can check each run and each figure on this page in the published run data.
The 62 runs are the normal referrals saved correctly in blocks 2 to 10, out of 63 new-patient referrals each. We left out block 1 because another process was competing for the CPU during it, and OpenAdapt is still the faster of the two when block 1 is counted. These runs compare speed and model cost only, on one test copy, and agent time depends on model latency on the day.
Technical details
With block 1 included, both saved 69 of 70. The medians over those 69 runs each were 29.7 s for OpenAdapt and 37.6 s for the agent. The same task from our earlier 10 Oct recording, compiled without the page's structure, saved 62 of 63 at a 38.6 s median, slower than the agent.
OpenAdapt's program was compiled with DOM element identity at engine commit 0c642bc0. The 10 Oct program was compiled without DOM element identity. The agent ran in our capped computer-use harness with hard cost and action caps.
Each OpenAdapt run that saved a chart followed the same 41 steps of this referral program without calling a model. The database read-back matched 12 of 12 fields of the new chart, and the runs that stopped saved nothing. In its correct runs the agent took a median of 49 actions and sent a median of 10 screenshots. These counts apply to this workflow on this test copy and not to OpenAdapt in general, because step and field counts depend on the form.
What OpenAdapt changed, as measured
- OpenAdapt's normal runs cost $0 in model fees per entry. The agent cost about $0.08 a run in the timed referral runs (median) and up to $0.89 in other tests.
- OpenAdapt's program ran the same 41 steps of this referral program in each timed referral run that saved a chart, while an agent plans each run as it goes.
- OpenAdapt sent no screens to a model provider in its recorded runs. A hosted agent sends a screenshot on each step, and Anthropic offers zero data retention for eligible organizations.
- The median time per entry was 29.7 s for OpenAdapt and 36.5 s for the agent. Both medians are over the 62 correct timed referral runs each.
- When a referral was a possible duplicate of an existing patient, neither OpenAdapt nor the agent made a new chart in 9 of 9 timed runs. The agent's result on the mistyped insured number isn't published, so the stop on a mismatch is only partly measured.
On simple tasks, the agent alone matched or beat the program in three of our tests, and matched it (62 of 63) in the timed referral runs. OpenAdapt helps when a task repeats, because the program follows the same steps without calling a model and reads each save back.
How your agent uses OpenAdapt
Your agent can do a new task the first time, and OpenAdapt compiles the successful run into a program so that later runs make no model calls. Your agent can also decide what to enter and call OpenAdapt to make the entry and check it. Both are released and tested on synthetic data.
Technical details
The agent connection is the OpenAdapt MCP server, started with `openadapt agent serve` (launcher 1.17.0, openadapt-agent 2.2.1). Compiling an agent's run into a program is the JIT path in openadapt-flow 2.2.0 (openadapt_flow.jit), tested on a recorded Claude run against a synthetic app.
See For AI agent teams.
Questions
Which model did you test?
We tested Claude Sonnet 5 (claude-sonnet-5) in July 2026 and Claude Sonnet 5.5 (claude-sonnet-5-5) in October 2026. Both are Anthropic computer-use models, and both ran at list price with hard cost and action caps in our harness.
Related: Browser agents · Claims intake · How it works · All comparisons
Sources
- OpenAdapt engine benchmark, agent arm, 21 Jul 2026 (README): a paid computer-use agent, Claude Sonnet 5 (claude-sonnet-5), on local test copies with synthetic data, each trial scored by a database read. openIMIS claim entry: 3 of 3 correct, $0.48 a run at list price, 69.8 s mean. Frappe Lending: 6 of 6 correct writes, 5 of 6 clean. Local OpenEMR: 0 of 6 (each trial used up its action budget before saving). Small samples; run on fresh baselines, not matched against OpenAdapt runs.
- OpenAdapt engine benchmarks, OpenEMR public demo, 8 Jul 2026 (docs/BENCHMARKS.md): an 18-step add-a-note task on the public demo (synthetic patients, resets daily). The compiled program passed 19 of 20 runs at 39.2 s median with 0 model calls; in the 20th the note stayed in the unsaved form. A computer-use agent, Claude Sonnet 5 (claude-sonnet-5), passed 10 of 10 at 70.4 s median, about $0.55 a run at list price. Both judged by the same screen check of the saved note, not a database read. Engine version 0.1.0.
- OpenAdapt on openIMIS 25.10, July 2026: one health-facility claim per run, 3 of 3 checked by a direct database read (exactly one claim, status Entered, right member), 0 duplicates, 25.6 to 30.3 s a run, 0 AI model calls.
- OpenAdapt on Frappe Lending 16.2.0 (open source), 16 Jul 2026, synthetic applicants: 6 of 6 loan applications checked through the API, a database read and a table-change audit (3 normal runs, 3 after a cosmetic screen change); 5 of 5 injected faults stopped before saving.
- OpenAdapt on OpenEMR 8.0.0.3 (open-source EMR), local test copy, 11 Oct 2026. One program ran over 8 synthetic eReferrals in one batch, with 0 AI model calls. 7 new patients were saved; our test setup read each chart back through the OpenEMR API and database, and 12 of 12 fields matched the referral. The 8th (Rosa Delgado, same name as an existing chart, different birth date) stopped at the EMR's duplicate check; afterwards no new chart existed. The engine labels the saved runs "Finished, not checked" (practice profile) and the stop "Check the record".
- OpenAdapt on openIMIS 25.10 (open-source insurance system), local test copy, 12 Oct 2026. A program learned from one demonstration entered 6 synthetic health-facility claims in one batch, with 0 AI model calls. 5 were saved; our test setup read each back through SQL and the openIMIS API, and 15 of 15 fields matched the claim document. On the 6th, the insured number had two digits swapped: openIMIS found no member, the screen no longer matched the demonstration, and the program stopped before saving (no claim was written). The engine labels the 5 saved runs "Finished, not checked", because this program has no record check of its own yet.
- Referral timing run, local OpenEMR 8.0.0.3 test copy, 14 to 15 Oct 2026 (run data): 8 synthetic referrals, OpenAdapt (a program compiled to find each field by the page's structure, 0 model calls) against Claude Sonnet 5.5 (claude-sonnet-5-5) as a computer-use agent in a capped harness, warm starts, interleaved blocks, each chart read back from the database. Blocks 2 to 10, normal referrals saved correctly: 62 runs each; median 29.7 s (OpenAdapt) and 36.5 s (agent); agent median $0.083 a run at list price. Both 62 of 63 normal referrals correct; on the possible duplicate, no new chart in 9 of 9 for both. Block 1 left out (CPU shared with another process); with it (blocks 1 to 10), 69 of 70 correct for both, 29.7 s and 37.6 s (n=69 each). The same task from the 10 Oct recording without the page's structure: 62 of 63, 38.6 s median, slower than the agent.
- Anthropic, Computer use tool documentation (platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool), and API data retention (platform.claude.com/docs/en/manage-claude/api-and-data-retention), read August 2026.
- OpenAI, Computer use guide (developers.openai.com/api/docs/guides/tools-computer-use), read August 2026.
See it on your process
Bring a task your team types every day. We'll tell you whether it fits and what a test on your own system would cover.