Report test progress for LLMs

LlmReporter is designed for use with Large Language Models (LLMs). It reports problems (warnings, skips, errors, and failures) as they occur and the total number of successes at the end.

LlmReporter is used by default when tests are run by a coding agent. Currently we detect Posit Assistant (via getOption("posit_assistant")), Claude Code, Codex, Cursor, and Gemini CLI (via env vars). If using another tool, configure it to set env var AGENT=1.

See also

Other reporters: CheckReporter, DebugReporter, FailReporter, JunitReporter, ListReporter, LocationReporter, MinimalReporter, MultiReporter, ProgressReporter, RStudioReporter, Reporter, SilentReporter, SlowReporter, StopReporter, SummaryReporter, TapReporter, TeamcityReporter