Get the default coding agent instructions. These are the immutable base instructions that cannot be modified by the agent. Long-term memory (

Extract the task name from a trial's config.json. Args: trial_dir: Path to the trial directory Returns: Task name if found,

Get agent_name for a thread.

리터럴 텍스트 검색 Codex 피드백 반영:** 정규식이 아닌 리터럴 문자열 검색 Python: grep_raw의 docstring - "검색할 리터럴 문자열 (정규식 아님)"

in-memory 파일 맵에서 구조화된 grep 매칭을 반환합니다. 성공 시 `list[GrepMatch]`를, 입력이 유효하지 않은 경우(예: 잘못된 정규식)는 오류 문자열을 반환합니다. 백엔드가 도구(tool) 컨텍스트에서 예외를 던지지 않도록 하고

Helper to check if Tavily API key is available

Hide the popup.

반복 횟수를 증가시키고 상태를 저장한다.

Check if the vertex is halted (can be reactivated)

재귀 한도 초과 확인

Check if this error is retryable

Check if the error is a timeout

List all superstep numbers (non-generic helper method)

Create LLM configuration

Create the LLM provider based on configuration Uses `RigAgentAdapter` to wrap Rig's native providers for full compatibility with rig-deepagents' midd

Set the maximum iterations for the explorer agent. Default: 5

메모리 콘텐츠를 system prompt에 주입합니다. Args: request: Model request to modify. Returns: Modified request with memory

모델 요청의 system prompt에 skills 섹션을 주입합니다. Args: request: Model request to modify Returns: New model request wi

Set the workflow name.

Create a new WebSearchTool with the specified max results. # Rust Concept: Associated Functions (Constructors) Functions that don't take `self` are

Prune checkpoints, keeping only the most recent `keep` checkpoints. This is useful for managing storage space in long-running workflows.

Replace text in the input from start to end with replacement. Args: start: Start index in the input text end: End ind

Reset token count.

네비게이션 상태를 초기화합니다.

Reset to a new thread. Returns the new thread_id.

Restore runtime state from a checkpoint # Critical Fixes (Gemini/Qwen Review) This method restores: - Vertex states (Active, Halted, Completed) - Pe

Run the workflow with the given initial state If the workflow was compiled with a checkpointer, checkpoints will be saved automatically at the config

Run the Textual application. Args: agent: Pre-configured LangGraph agent (optional) assistant_id: Agent identifier for memory sto

Get the underlying runtime for advanced configuration

Scan a directory for SKILL.md files

Set whether completion suggestions are visible.

Set whether the cursor should be actively blinking. When active=False (e.g., agent is working), disables cursor blink so the cursor d

Set text from history navigation.

Check if the workflow should terminate

현재 상태 참조

Stop the streaming and finalize the content.

Generate a Mermaid diagram with node kinds for proper shape rendering. Use this when you have NodeKind information available (e.g., from a WorkflowGr

Generate a Mermaid diagram with both state colors and node shapes.

도구 호출 ID

Trim messages to fit within the summarizer's token budget.

여러 파일을 state로 업로드합니다.

Mark directions as explored

Include a general-purpose agent

Set max iterations for subagent execution

Arc로 래핑된 미들웨어 추가

Set the model name

Enable parallel direction exploration.

Set retry policy

Add a snippet to the source

Set custom system prompt

Add token usage statistics to the response

Set workflow timeout

Write initial content if provided at construction time.

Create the approval menu widget for a single action request.

파괴적인 도구에 대한 HITL(Human-In-The-Loop) interrupt_on 설정을 구성합니다.

히스토리 파일에 항목 하나를 append 합니다(concurrent-safe).

Apply the currently selected completion.

소스 유형의 다양성을 계산한다. Returns: 0.0 ~ 1.0 범위의 다양성 점수 (4종류 이상이면 1.0).

Clean up after agent task completes or is cancelled.

Clear the messages area.

오래된 항목을 제거하기 위해 히스토리 파일을 재작성합니다. Only called when entries exceed 2x max_entries to minimize rewrites.

Compute additions and deletions count. Returns: Tuple of (additions, deletions)

Count additions and deletions from diff data.

Create a new skill with a template SKILL.md file. Args: skill_name: Name of the skill to create. agent: Agent identifier for skil

연구 에이전트 생성.

Create a LangSmith experiment session. Args: dataset_id: LangSmith dataset ID to associate with name: Name for the experiment ses

Glob 도구를 생성합니다.

Grep 도구를 생성합니다.

read_file 도구를 생성합니다.

서브에이전트를 호출하는 `task` 도구를 생성합니다. Args: default_model: Default model for subagents. default_tools: Default tools for subagents.

검색 도구들을 생성합니다.

Auto-detect provider from model name. Args: model_name: Model name to detect provider from Returns: Provider name (openai, a

Docker 사용 가능 여부를 확인합니다.