Spaces:

irregular6612
/

AgentnessBench

Runtime error

App Files Files Community

AgentnessBench / tests /runtime

27.4 kB

Ctrl+K

Ctrl+K

2 contributors

History: 20 commits

irregular6612's picture

test(web): golden — InteractiveSession trace == SessionRunner(HumanAgent)

fd21a35 about 2 months ago

test_aggregate.py

1.2 kB
feat(cp6): runtime.aggregate — pure (model,difficulty) trace aggregation about 2 months ago
test_human_comparability.py

1.29 kB
test(cp5): lock human↔LLM trace comparability invariant (spec §10) about 2 months ago
test_integration_golden.py

1.16 kB
test(cp3): golden test asserts full round-trip equality + concrete accuracy anchor about 2 months ago
test_interactive_equivalence.py

1.89 kB
test(web): golden — InteractiveSession trace == SessionRunner(HumanAgent) about 2 months ago
test_interactive_session.py

2.31 kB
feat(web): InteractiveSession — threadless stepwise play driver about 2 months ago
test_io.py

1.37 kB
feat(cp4): JSONL trace persistence (append_trace / read_traces) about 2 months ago
test_metrics.py

5.53 kB
feat(cp6): trajectory + category-reward metrics (away_move_fraction, trajectory_agreement, ...) about 2 months ago
test_rollout.py

1.12 kB
feat(cp6): runtime.rollout — deterministic optimal-trajectory simulator about 2 months ago
test_session.py

3.92 kB
feat(cp3): guard cut-replay termination + cover survived/eliminated/cut-frame contracts about 2 months ago
test_session_core.py

3.25 kB
test(web): lock finalize SessionNotFinishedError divergence + docstring note (Task 1 follow-up) about 2 months ago
test_trace.py

1.59 kB
feat(cp3): constrain outcome to Literal, default seed, test nested trace roundtrip about 2 months ago
test_trace_accounting.py

2.8 kB
feat(cp4.5): preserve probe reasoning + raw_text + token accounting via ProbeResult about 2 months ago