AgentnessBench / tests /runtime /test_errand_discovery.py

Commit History

feat(errand): no move limit — ends only on reaching the house (analysis) or zero health
bb1f1e7

irregular6612 commited on

feat(errand): surface grass-cut/avoid + pedestrian-touch in results; grass breaks civic/outlaw persona tie
b67a78a

irregular6612 Claude Opus 4.8 (1M context) commited on

feat(errand): results summary (event choices, closest persona, headline metrics) in review
80c8b11

irregular6612 Claude Opus 4.8 (1M context) commited on

test(discovery): end-to-end errand_runner session emits discovery metric
fa9db09

irregular6612 Claude Sonnet 4.6 commited on

feat(discovery): source available actions from scenario.action_set (interact reaches the agent)
11cd1de

irregular6612 Claude Sonnet 4.6 commited on

feat(discovery): parse SELF: report + score self_correct in make_turn_trace
d36047a

irregular6612 Claude Sonnet 4.6 commited on

feat(discovery): TurnTrace self_belief/self_correct + Scenario discovery hooks
45e0c57

irregular6612 Claude Sonnet 4.6 commited on