Skip to main content
Momentic uses several specialized AI agents. Each is versioned independently, so you can upgrade one without changing the others. Older versions stay available for backwards compatibility; set a specific version per agent in your momentic.config.yaml ai.agentConfig block. Momentic announces agent deprecations at least 90 days in advance.
When upgrading an agent version, we recommend testing the migration in a branch first. AI model decisions can vary between versions.
Supported locator and assertion agents use memory (insights saved from past runs) and custom knowledge base entries. Agents can fall back to configured alternative models when a provider fails. Fallbacks can change latency and decisions. Memory, the knowledge base, and failure recovery reduce the frequency and impact of AI non-determinism on your tests.

Web agents

Mobile agents

V5 effort routing

V5 requires momentic 3.52.10 or later. v5 routes each locator and assertion query to one of three effort levels (fast, medium, or extended) based on the query’s complexity. You set v5 in agentConfig; the effort level is not configurable. The effort level multiplies the AI step rate for the query: Multipliers apply to usage-based plans: Free, Scale, and older Free and Pay-as-you-go plans, where they multiply the usage a query bills. Enterprise contracts priced per test are not affected. visual-assertion v5 does not use effort routing. Every visual assertion bills at the 1x AI step rate, the same as v4. Setting visual-assertion: v5 in momentic.config.yaml requires momentic 3.60.0 or later; selecting it in workspace settings works with any CLI version.

What each agent does

Locates elements from a natural language description, used by Click, Type, and Element check steps. Exact single-quoted text constrains matching. Successful resolutions can populate the step cache.
Evaluates natural language statements against a snapshot of the page, used by AI check steps. Prioritizes exact single-quoted text values and resolves visual detail from the current page context.
Evaluates natural language statements from a viewport screenshot. Use it for visible layout, text, and appearance within the viewport.
Extracts structured data from the page given a JSON schema, used by AI extract steps. Supports nested objects and arrays.
Generates and executes recovery steps when a recoverable failure is detected. Requires ai.failureRecovery.
Locates elements in the Android UI hierarchy from a natural language description, used by mobile Tap, Type, and Element check steps.
Locates elements in the iOS UI hierarchy from a natural language description, used by mobile Tap, Type, and Element check steps.