Details

---

Object Hierarchy:

Description:

Content:

Properties:

• public ListModel children { get; }
  **Execution queue** — build_run_queue / run_exec (activity). May reuse proposed_tools rows or add exam-only / synthetic exec rows. Same property as ProgressItem.children; default instance is ToolList.
• public ArrayList<Block> code_blocks { get; set; }
  Code blocks from refinement (parser); add directly.
• public ArrayList<Format> exam_references { get; set; }
  Links from "examination references" task_data (per-run execution).
• public bool exec_done { get; set; }
  True after run_exec success; children then hold summaries and documents.
• public ulong idx_notify_id { get; set; }
• public string issues { get; set; }
  Validation errors; append with this.issues += "\n" + msg. Parser checks and appends with section context.
• public bool last_failure_was_communication { get; }
  True when refinement failed after exhausting communication retries (e.g. send threw 3 times). Caller should report to user.
• public Message? message { get; set; }
• public string msg_idx_txt { owned get; }
• public Document out_doc { get; set; }
  Single markdown document after post-exec synthesis. Headings used for

   task://slug.md 

  (full document) or optional fragment for one section. Empty until run_post_exec finishes.
• public Block post_summary { get; set; }
  Result summary block from post-exec (iteration + completed-task list).
• public ToolList proposed_tools { get; set; }
  Refined tool-call **proposals** ( ResultParser / to_markdown). **Not** nested progress rows — see children.
• public ArrayList<Format> references { get; set; }
  Markdown links from task "references" block (current_file, paths, plan:...); Runner resolves for prompt fill. Filled in fill_task_data().
• public bool requires_user_approval { get; set; }
  True when this task should gate execution (e.g. modifies files); from task list format.
• public ResultParser result_parser { get; set; }
  Parser for last refine or executor response; children valid after exec_done (from the execution queue). Initialized in ctor so issues is always available.
• public Runner runner { get; }
  Runner; looked up from step.list.runner (tree: runner → list → step → details).
• public ArrayList<Format> shared_references { get; set; }
  Links from "shared references" task_data (refinement).
• public Definition skill { get; set; }
  Resolved YAML skill for this task.
• public Manager skill_manager { get; }
  Alias to runner.sr_factory.skill_manager (no setter).
• public PhaseEnum status { get; set; }
• public string status_str { owned get; }
• public unowned Step step { get; set; }
  Step this task belongs to; required so Details looks up the tree (step.list.runner) for runner, session, etc.
• public int step_index { get; set; }
  Step index (0-based) of the section this task belongs to, or -1 if not set. Set when the task list is built (e.g. in ResultParser.parse_task_list) so we don't re-derive section from position later.
• public Map<string,Block> task_data { get; set; }
  Map from ListItem.to_key_map(..., a_2_z); keys are lowercase labels only. All task content is read from here.
• public string title { owned get; }
• public RequestBase? tool_request { get; set; }
• public string tooltip_text { owned get; }

Creation methods:

• public Details (Step step, Map<string,Block> task_data)
  Initial task_data from ListItem.to_key_map(..., a_2_z) for one task list item (keys: lowercase labels only, see class doc). Runner and session come from step.list.runner (tree).

Methods:

• public void build_run_queue ()
• protected override async void fill_model ()
  Set chat_call.model from this task's skill definition when the skill header has an optional model and it is available; otherwise use default.
• public void fill_name (int i)
  If task_data has no "name" or empty, set name = (skill or "Task") + " " + index. Single method.
• public string issue_label ()
  Label for this task in issue messages: section number and name or slug (e.g. "section 2 \"Research 1\"" or "section 2 (slug: research-1)") so the LLM can locate the task in the task list.
• public PromptTemplate post_exec_prompt (string previous_response, string retry_issues) throws Error
  Build post-execution prompt from task_post_exec.md. previous_response and retry_issues are used for retries (header_raw when non-empty).
• public async void refine (Cancellable? cancellable = null) throws Error
  Refinement: fill template. Caller has validated via skill_manager.validate(this); definition from this.skill is non-null. task_reference_contents comes from reference_contents() → ResolveLink after preload_links (this.references only) in this loop. Up to 5 refinement attempts; up to 3 communication retries per attempt. Caller (Runner) must catch and report to user; see 1.23.14.
• public PromptTemplate refinement_prompt () throws Error
  Build refinement prompt template (no current_file; reference_contents includes current file when in the task's References).
• public async void run_exec () throws Error
  Run all Tool exec runs (tool if needed, then LLM) in order — every run in the execution queue (tools). Then post-exec synthesis when there is more than one run. Summaries and canonical document from post-exec when applicable.
• public async void run_post_exec () throws Error
  Post-execution synthesis: combine executor outputs, call LLM with task_post_exec.md, parse into post_summary and out_doc; validate links. Retry on parse/validation issues: refill with previous output and issues via header_raw (same pattern as refinement).
• public string slug ()
  This task's name as slug (e.g. "Research 1" → "research-1"). Returns "" if no name in task_data or empty.
• public string to_markdown (PhaseEnum phase)
  Task as markdown for a given phase. Does not add section headings (e.g. `## Task`); caller adds header. COARSE: creation keys. REFINEMENT/EXECUTION: task list + `## Tool Calls` when tools exist. POST_EXEC: task list only (no Tool Calls). LIST: task list + ##### Result summary when exec_done.
• public string tool_instructions (Definition definition)
  Fenced JSON blocks from the session tool registry for each name in the skill's tools list. Empty when there are no tools, the skill lists write_file, or nothing resolves to BaseTool.
• public ToolList tools ()
  Typed view of children — optional sugar instead of casting at every call site.
• public void update_props (Map<string,Block> refined_map)
  Apply refined task map: set each key from refined_map into this.task_data, then re-run fill.
• public async void wait_refined () throws Error
• public void write (string suffix = "")
  Write this task's result to session task dir. Writes a single slug.md from out_doc.

Inherited Members: