Agent classes / Computer-use agent / Form-fill RPA
Form-fill RPA for a computer-use agent
Updated 2026-10-27 (newest stored date on this page) · score asOf 2026-09-14 (stale)
The class: Operates a desktop or mobile GUI, not only a single site API, and checks that the control it used is the one it intended.
- A screenshot or accessibility tree after each action
- An allowlist of apps the task may change
Depot Pick: Skyvern
When not to use it: Need stealth multi-session browsers without RPA opinionation (Anchor/Browserbase), or open-source self-host Browser Use.
Runner-up: Anchor Browser.
No single-unit sourced price comparison for this job (no_comparable_unit_for_job).
Failure modes this class and job share (7)
- Benchmark protocol mismatch: Store assisted-or-not, subset, and harness version beside every cited score. Refuse a ranking that mixes those fields.
- Collateral side effects: Snapshot the workspace or app state, then diff it against an allowlist. A passing check plus an unexpected diff is a failure.
- Confirmation skip: Gate those tool names on an approval id recorded before the call. A missing approval is a failure even if the downstream tool returns success.
- GUI grounding miss: After each GUI action, compare the resulting screen or accessibility tree to the control the plan named. Stop when they differ.
- Partial tool success: Require a per-item status and a failure count. Success for the task requires the failed count to be zero or explicitly handed to a retry queue.
- Prompt injection: Compare each tool call to the original user task. If the call only makes sense given text that arrived from a fetched page, file, or message, stop and ask. Record which source the instruction came from. This catalog does not store payloads or attack steps.
- Tool loop: Hash the last several tool names and arguments. Stop when the same call repeats and the observation did not change. Cap tool calls per task.
Free options this class uses (0)
No free row lists this job with a kind this class needs.