Jesse 528b3cfe5a

feat: configurable context budget for tool-loop iterations (#2317) ...

* feat: add contextBudgetTokens config field for tool-loop trimming

* feat: implement _trim_history_for_budget for tool-loop cost reduction

* feat: thread contextBudgetTokens into AgentLoop constructor

* feat: wire context budget trimming into agent loop

* refactor: move trim_history_for_budget to helpers and add docs

- Extract trim_history_for_budget() as a pure function in helpers.py
- AgentLoop._trim_history_for_budget becomes a thin wrapper
- Add docs/CONTEXT_BUDGET.md with usage guide and trade-off notes
- Replace wrapper tests with direct helper unit tests

---------

Co-authored-by: chengyongru <chengyongru.ai@gmail.com>

2026-03-23 18:13:03 +08:00

..

test_azure_openai_provider.py

fix(azure): sanitize messages and handle temperature

2026-03-07 03:57:57 +00:00

test_base_channel.py

fix(auth): prevent allowlist bypass via sender_id token splitting

2026-03-07 16:36:12 +00:00

test_channel_plugins.py

fix: cross-platform test compatibility

2026-03-14 16:13:38 +08:00

test_cli_input.py

feat(cli): extract streaming renderer to stream.py with Rich Live

2026-03-23 10:20:41 +08:00

test_commands.py

refactor(agent): unify process_direct to return OutboundMessage

2026-03-22 00:39:38 +08:00

test_config_migration.py

refactor: remove deprecated memory_window, harden wizard display

2026-03-20 18:46:13 +08:00

test_config_paths.py

refactor: unify instance path resolution and preserve workspace override

2026-03-08 02:58:25 +00:00

test_consolidate_offset.py

refactor: remove deprecated memory_window, harden wizard display

2026-03-20 18:46:13 +08:00

test_context_multimodal.py

Make multimodal input limits configurable

2026-03-22 20:45:56 +08:00

test_context_prompt_cache.py

fix: remove stale IDENTITY bootstrap entry

2026-03-08 15:39:40 +00:00

test_cron_service.py

feat(cron): add run history tracking for cron jobs

2026-03-21 02:28:35 +08:00

test_cron_tool_list.py

fix(cron): preserve exact intervals in list output

2026-03-18 12:39:06 +08:00

test_cron_tool_tz_at.py

fix(cron): support tz parameter with at for one-time scheduled tasks

2026-03-22 21:00:42 +08:00

test_custom_provider.py

test(custom): cover empty choices response handling

2026-03-17 14:24:55 +08:00

test_dingtalk_channel.py

fix: save DingTalk downloaded files to media dir instead of /tmp

2026-03-15 23:21:22 +08:00

test_docker.sh

docs: add core agent line count script and update README with real-time stats

2026-02-06 07:28:39 +00:00

test_email_channel.py

fix(email): preserve fetched messages across IMAP retry

2026-03-21 03:00:39 +08:00

test_evaluator.py

refactor: replace <SILENT_OK> with structured post-run evaluation

2026-03-14 17:41:08 +08:00

test_exec_security.py

security: add SSRF protection, untrusted content marking, and internal URL blocking

2026-03-16 15:05:26 +08:00

test_feishu_markdown_rendering.py

test(feishu): cover heading and table markdown rendering

2026-03-17 16:51:02 +08:00

test_feishu_post_content.py

Merge main into pr-332

2026-03-07 15:02:06 +00:00

test_feishu_reply.py

test(feishu): cover media msg_type mapping

2026-03-17 17:05:13 +08:00

test_feishu_table_split.py

fix(feishu): split card messages when content has multiple tables

2026-03-01 15:13:44 +01:00

test_feishu_tool_hint_code_block.py

fix(feishu): avoid breaking tool hint formatting and think stripping

2026-03-14 15:40:53 +00:00

test_filesystem_tools.py

fix(agent): harden multimodal tool result flow

2026-03-21 05:34:56 +00:00

test_gemini_thought_signature.py

refactor: centralize tool call serialization in ToolCallRequest

2026-03-11 15:32:43 +08:00

test_heartbeat_service.py

fix(heartbeat): inject shared current time context into phase 1

2026-03-16 10:52:26 +08:00

test_litellm_kwargs.py

fix(openrouter): revert custom_llm_provider, always apply gateway prefix

2026-03-15 10:52:36 +08:00

test_loop_consolidation_tokens.py

fix(memory): reserve completion headroom for consolidation

2026-03-23 11:54:44 +08:00

test_loop_save_turn.py

fix: preserve image paths in fallback and session history

2026-03-17 22:37:09 +08:00

test_matrix_channel.py

feat: channel plugin architecture with decoupled configs

2026-03-14 16:13:38 +08:00

test_mcp_tool.py

fix(mcp): preserve schema semantics during normalization

2026-03-21 14:35:47 +08:00

test_memory_consolidation_types.py

fix(memory): validate save_memory payload and raw-archive on repeated failure

2026-03-13 03:53:50 +00:00

test_message_tool_suppress.py

refactor(memory): switch consolidation to token-based context windows

2026-03-10 19:55:06 +00:00

test_message_tool.py

test: remove message tool media test (message.py changes out of scope)

2026-02-20 18:06:13 +01:00

test_mistral_provider.py

feat(providers): add Mistral AI provider

2026-03-23 11:07:46 +08:00

test_onboard_logic.py

refactor(providers): hide oauth-only providers from config setup

2026-03-21 03:20:59 +08:00

test_provider_retry.py

fix: preserve image paths in fallback and session history

2026-03-17 22:37:09 +08:00

test_providers_init.py

fix(providers): lazy-load provider exports

2026-03-18 22:01:29 +08:00

test_qq_channel.py

fix(qq): fix local file outbound and add svg as image type (#2294)

2026-03-22 20:45:31 +08:00

test_restart_command.py

refactor(agent): unify process_direct to return OutboundMessage

2026-03-22 00:39:38 +08:00

test_security_network.py

security: add SSRF protection, untrusted content marking, and internal URL blocking

2026-03-16 15:05:26 +08:00

test_session_manager_history.py

fix: keep truncated session history tool-call consistent

2026-03-16 17:25:30 +08:00

test_skill_creator_scripts.py

fix(skill-creator): restore validation and align packaging docs

2026-03-10 09:16:23 +00:00

test_slack_channel.py

feat(slack): add default done reaction on completion

2026-03-17 16:19:08 +08:00

test_task_cancel.py

refactor(tools): keep exec enable without configurable deny patterns

2026-03-20 17:46:08 +00:00

test_telegram_channel.py

fix(agent): make status command responsive and accurate

2026-03-21 15:21:32 +00:00

test_tool_validation.py

fix(mcp): preserve schema semantics during normalization

2026-03-21 14:35:47 +08:00

test_trim_history_for_budget.py

feat: configurable context budget for tool-loop iterations (#2317)

2026-03-23 18:13:03 +08:00

test_web_fetch_security.py

fix(agent): harden multimodal tool result flow

2026-03-21 05:34:56 +00:00

test_web_search_tool.py

feat(web): multi-provider web search + Jina Reader fetch

2026-03-13 05:44:16 +00:00