NCBM 86596633e8 core/agent: confirm a round's mutating calls in one prompt ...

ToolRegistry.prepare_batch_confirm pre-scans a round of tool calls, drops the ones that cannot run (unknown tool, undecodable arguments, missing required argument), and turns two or more soft reasons into a single ToolBatchConfirmHook prompt. The BatchConfirmDecision it returns is applied per call through execute(..., decision=...): allow, or deny with one comment shared by every call of that round.

_execute_tool_calls asks before it gathers, so a batch of moves or deletes no longer stacks one confirm box per call. A single request keeps the old one-call prompt, a host without a batch hook keeps the per-call behaviour, and the answer never outlives the round it was asked in.

2026-10-17 17:32:44 +08:00

..

test_budget.py

core/agent: token-ify generic tool-result cap and compact shrinks

2026-08-11 19:41:54 +08:00

test_compact_summarize.py

core/agent: tell the compact summarizer that tools are unavailable

2026-10-17 17:13:42 +08:00

test_compact.py

core/agent: token-ify generic tool-result cap and compact shrinks

2026-08-11 19:41:54 +08:00

test_confirm.py

test/tools: catalog, grants, and async tool helpers

2026-07-24 14:57:03 +08:00

test_directive_checkpoint.py

core/agent: append-only directive checkpoints on token bands

2026-07-24 19:40:37 +08:00

test_finish_reason.py

core/agent: unify provider finish_reason mapping

2026-08-01 01:05:26 +08:00

test_loop.py

core/agent+cli: keep a round cut off at the output limit

2026-10-16 22:29:25 +08:00

test_messages_bridge.py

core/agent+anthropic: drop the thinking block a model refuses to disable

2026-10-16 22:16:25 +08:00

test_messages_dispatch.py

core/agent+anthropic: drop the thinking block a model refuses to disable

2026-10-16 22:16:25 +08:00

test_notices.py

core/agent+cli: keep a round cut off at the output limit

2026-10-16 22:29:25 +08:00

test_parallel_tools.py

core/agent+tools: apply a batch's same-file writes in call order

2026-10-16 19:37:29 +08:00

test_responses_bridge.py

core/agent: return each round's reasoning in the Responses input

2026-10-06 23:23:18 +08:00

test_responses_dispatch.py

core/agent: stream a repeated reasoning text only once

2026-10-07 00:26:38 +08:00

test_retry_nag_integration.py

core/agent: remove turn-start todo nag -- nag only fires after complete turn

2026-07-29 01:07:59 +08:00

test_retry_todo_nag.py

lint: drop dead code that blocks basedpyright (unused helper and import)

2026-08-01 00:33:25 +08:00

test_rollback.py

core/agent: keep a rolled-back turn's reasoning with its calls

2026-10-06 23:23:12 +08:00

test_run_aside.py

cli/btw: tell the model side questions are unsaved and tool-scoped

2026-09-19 09:24:38 +08:00

test_stream_merge.py

core/lmproto+agent: use normal stream API for chat loop

2026-07-14 22:40:15 +08:00

test_todo_stack.py

core/agent+tools+cli: number todos by listing order, not the internal id

2026-10-16 18:52:36 +08:00

test_tools_batch_confirm.py

core/agent: confirm a round's mutating calls in one prompt

2026-10-17 17:32:44 +08:00

test_tools.py

agent/tools: report missing required arguments before running

2026-09-19 09:50:13 +08:00

test_usage.py

core/agent: normalize provider usage structs in token_usage_from_api

2026-08-01 01:09:12 +08:00