Skip to contents

Returns the small set of health numbers behind an agent_run: call counts, token totals, tool and blocked-event counts, wall-clock duration, and the study's manifest_hash. Token and outcome figures are read through LLMR::llm_usage() over the run's call-level rows, so they match the rest of the LLMR ecosystem to the integer.

Usage

# S3 method for class 'agent_run'
diagnostics(x, ...)

Arguments

x

An object accepted by as_agent_run() (an Agent, a conversation or preset result, a pipeline, an experiment, or an agent_run).

...

Unused.

Value

A one-row tibble with run_id, kind, n_calls, n_ok, n_failed, ok_rate, n_truncated, n_filtered, tokens_sent, tokens_received, tokens_total, reasoning_tokens, n_tool_calls, n_blocked, duration_s, and manifest_hash.

Examples

if (FALSE) { # \dontrun{
a <- agent("Aria", LLMR::llm_config("groq", "openai/gpt-oss-20b"))
a$chat("Hello")
diagnostics(a)
} # }