deepseek-harness/packages/core/agent-loop/tests
_Kerman ab56e0df0a fix(agent-loop): park empty turn batches, keep max-tokens sticky, and settle headless at idle
- turn boundary now returns false for an empty admitted batch (claimed
  input removed before the wake) instead of opening a turn and spending a
  model call on nothing; the step boundary already had the symmetric guard.
- a max-token step stays sticky when steering or injected work continues
  the turn: a later completed step no longer downgrades the outcome,
  matching the TurnEndReasonMap contract.
- session/queue wire schema accepts the context placement (previously the
  zod union rejected injected-context snapshots wholesale and the client
  silently dropped the whole frame); schema tests cover all placements.
- headless runs settle at whole-agent idle instead of the first turn/end,
  honoring the one-shot idle-to-idle contract.
- flush JSDoc names the real callers (checkpoint policy, goal-session,
  teardown, self-flushing consumers); apiproxy zh README loses its stale
  duplicate history section; ACP note/README record the delivered error
  rejection and turnless-cancelled behaviors.
2026-08-03 20:36:26 +08:00
..
agent-initiator.spec.ts refactor agent pre-step inbox lifecycle 2026-07-31 19:21:16 +08:00
agent.spec.ts fix(subagent): confirm steering request admission 2026-08-02 04:34:16 +08:00
cancel.spec.ts refactor(agent-loop): drop the steering/message session event 2026-08-03 16:04:26 +08:00
config-session-id.spec.ts Merge branch 'master' of https://github.com/deepseek-harness/deepseek-harness into xtr/react-loop-simplification 2026-07-31 10:16:14 +08:00
contract-regressions.spec.ts fix(agent-loop): park empty turn batches, keep max-tokens sticky, and settle headless at idle 2026-08-03 20:36:26 +08:00
coverage-edges.spec.ts fix(session): type turn/end error as one structured failure 2026-08-03 16:33:23 +08:00
interception.spec.ts refactor(agent-loop): drop the steering/message session event 2026-08-03 16:04:26 +08:00
invariant.spec.ts fix: commit step context before request dispatch 2026-07-30 18:18:04 +08:00
loop.spec.ts fix(agent-loop): park empty turn batches, keep max-tokens sticky, and settle headless at idle 2026-08-03 20:36:26 +08:00
mock-adapter.ts feat(llm-deepseek): configure max token defaults 2026-07-30 21:04:00 +08:00
properties.spec.ts test: align consumers with owned-run semantics 2026-07-30 17:45:58 +08:00
request-cache.e2e.ts refactor(llm-deepseek)!: rename the provider route to deepseek-official 2026-07-29 16:36:07 +08:00
request-error.spec.ts test: align consumers with owned-run semantics 2026-07-30 17:45:58 +08:00
request-reconstruction.spec.ts fix(session): type turn/end error as one structured failure 2026-08-03 16:33:23 +08:00
resume.spec.ts Merge remote-tracking branch 'origin/master' into xtr/react-loop-simplification 2026-08-03 10:56:09 +08:00
runtime-context.spec.ts refactor(agent): complete inbox lifecycle migration 2026-08-03 12:25:33 +08:00
scope-lifecycle.spec.ts Merge remote-tracking branch 'origin/master' into xtr/react-loop-simplification 2026-08-03 10:56:09 +08:00
tool-calls.spec.ts fix(session): type turn/end error as one structured failure 2026-08-03 16:33:23 +08:00
tool-order.spec.ts refactor(agent-loop): project runtime context before steps 2026-08-01 20:39:10 +08:00