dominds 0.7.6 → 0.8.2

package/dist/docs/dominds-agent-priming.zh.md

@@ -4,7 +4,7 @@
 
 ## 摘要
 
-Dominds 的诉请（Tellask，`!?@...`）与扪心自问（FBR，`!?@self`）并不是“写在系统提示里的约定”，而是**后端真实可驱动、可执行、可回传、可持久化**的机制。
+Dominds 的诉请（Tellask，`tellask* function call`）与扪心自问（FBR，`freshBootsReasoning`）并不是“写在系统提示里的约定”，而是**后端真实可驱动、可执行、可回传、可持久化**的机制。
 
 问题在于：即使系统提示把这些机制写得再细，大多数基础模型也并未在“真的可以随时拉一个队友去跑 shell 命令并把结果带回来”的世界里被训练过，因此容易把提示当成“说说而已 / 仅供参考”。这会直接削弱诉请与 FBR 的可用性。
 
@@ -20,7 +20,7 @@ Dominds 的诉请（Tellask，`!?@...`）与扪心自问（FBR，`!?@self`）并
 
 - 诉请机制与对话系统：[`dialog-system.zh.md`](./dialog-system.zh.md)
 - 术语约定（主线/支线；诉请方/应答方）：[`dominds-terminology.md`](./dominds-terminology.md)
-- FBR（`!?@self`）：[`fbr.zh.md`](./fbr.zh.md)
+- FBR（`freshBootsReasoning`）：[`fbr.zh.md`](./fbr.zh.md)
 - 工作语言 vs UI 语言：[`i18n.zh.md`](./i18n.zh.md)
 
 ---
@@ -28,7 +28,7 @@ Dominds 的诉请（Tellask，`!?@...`）与扪心自问（FBR，`!?@self`）并
 ## 目标
 
 - 在对话一开始就建立“诉请/回传/持久化都是真的”的信任。
-- 在对话一开始就跑通一次 `!?@self` 的 FBR 回路。
+- 在对话一开始就跑通一次 `freshBootsReasoning` 的 FBR 回路。
 - 把“发起 FBR 后先等待回贴，再做综合决策”的时序固化为肌肉记忆。
 - 把多份 FBR 草稿做一次**综合提炼**，把“取精华、去糟粕”的动作也变成体感的一部分。
 - 过程足够小、可控、低风险（默认只执行 `uname -a`）。
@@ -46,9 +46,9 @@ Dominds 的诉请（Tellask，`!?@...`）与扪心自问（FBR，`!?@self`）并
 
 - **主线对话**：用户与主要智能体交互的那条线。
 - **支线对话**：由诉请 / FBR 触发的临时工作线，产出结果回传主线。
-- **诉请（Tellask）**：以 `!?@<memberId> ...` 向队友/对话发起的结构化请求。
+- **诉请（Tellask）**：以 `tellask({ targetAgentId: "<memberId>", sessionSlug: "<slug>", tellaskContent: "..." })` 向队友/对话发起的结构化请求。
 - **Shell 专员（shell specialist）**：被允许执行 shell 命令并回传结果的队友（由 `shell_specialists` 配置指定）。
-- **FBR（扪心自问）**：以 `!?@self` 触发的“无工具支线推理”，产出报告回传主线。
+- **FBR（扪心自问）**：以 `freshBootsReasoning` 触发的“无工具支线推理”，产出报告回传主线。
 
 ---
 
@@ -83,7 +83,7 @@ Dominds 的诉请（Tellask，`!?@...`）与扪心自问（FBR，`!?@self`）并
 
 ### 3）真实 FBR：基于环境信息做一次“扪心自问”
 
-拿到环境快照后，主线智能体触发一次 `!?@self` FBR，并在 FBR 的诉请正文里带上：
+拿到环境快照后，主线智能体触发一次 `freshBootsReasoning` FBR，并在 FBR 的诉请正文里带上：
 
 - 命令 `uname -a` 的完整输出（明确注明命令名，未来可能更换命令，不能靠猜）
 - FBR 的工具约束（FBR 无工具；主线工具另论）
@@ -91,14 +91,14 @@ Dominds 的诉请（Tellask，`!?@...`）与扪心自问（FBR，`!?@self`）并
 
 并发草稿（可选）：
 
-- 若团队成员配置启用了 `fbr_effort`（默认值为 `3`），运行时会并发创建多份 `!?@self` FBR 支线对话，让“初心自我”从不同角度独立推理，形成多份草稿供主线对话综合提炼。
+- 若团队成员配置启用了 `fbr_effort`（默认值为 `3`），运行时会并发创建多份 `freshBootsReasoning` FBR 支线对话，让“初心自我”从不同角度独立推理，形成多份草稿供主线对话综合提炼。
 - 这些草稿之间**没有稳定身份映射**，也没有必须遵循的先后顺序；主线对话应当把它们当作“多份匿名草稿”，而不是“固定的三个角色”。
 - 若 `fbr_effort` 为 `0`，则跳过 FBR。
 - 若 `fbr_effort` 大于 `100`，运行时会报错并终止启动流程（配置错误）。
 
 阶段边界（关键）：
 
-- `!?@self` 只是**发起动作**，不是“已完成决策”。
+- `freshBootsReasoning` 只是**发起动作**，不是“已完成决策”。
 - 主线对话必须先进入等待态，直到该次 FBR 的回贴返回。
 - 若 `fbr_effort = N`，主线必须等待全部 N 条回贴后再进入综合提炼；不得基于部分草稿提前定稿。
 
@@ -175,7 +175,7 @@ Dominds 的诉请（Tellask，`!?@...`）与扪心自问（FBR，`!?@self`）并
 - 智能体启动应以真实对话转录展示，用户可展开查看细节。
 - 建议在对话顶部提供一个可折叠区域，并带上清晰标签：
   - “队友诉请（shell）”
-  - “FBR（`!?@self`）”
+  - “FBR（`freshBootsReasoning`）”
   - “智能体启动（Agent Priming）”
 
 ### 创建对话时可选择跳过（opt-out）

package/dist/docs/dominds-terminology.md

@@ -42,7 +42,7 @@
 - EN: `Sideline dialog` | ZH: `支线对话`
 - EN: `Taskdoc` | ZH: `差遣牒`
 - EN: `Taskdoc package (*.tsk/)` | ZH: `任务包`
-- EN: `!tellaskSession <slug>` | ZH: 会话 Slug（只写在 headline）
+- EN: `sessionSlug` | ZH: 会话 Slug（只写在 headline）
 - EN: `CLI (entrypoint UI)` | ZH: `CLI（入口界面）`
 - EN: `TUI (interactive UI)` | ZH: `TUI（交互前端）`
 - EN: `WebUI (interactive UI)` | ZH: `WebUI（交互前端）`
@@ -82,19 +82,19 @@
 
 #### Tellask headline（诉请头）
 
-- EN: The first line of a Tellask block, starting with `!?@<name> ...`.
-- ZH: 诉请块的第一行，以 `!?@<name> ...` 开头。
+- EN: The first line of a Tellask block, starting with `tellaskSessionless({ targetAgentId: "<name>", tellaskContent: "..." })`.
+- ZH: 诉请块的第一行，以 `tellaskSessionless({ targetAgentId: "<name>", tellaskContent: "..." })` 开头。
 
-- EN: Additional lines starting with `!?@...` in the same block are appended to the headline (they do not start a new Tellask).
-- ZH: 同一诉请块内，后续以 `!?@...` 开头的行会被并入诉请头（不会开启新的诉请）。
+- EN: Additional lines starting with `tellask* function call` in the same block are appended to the headline (they do not start a new Tellask).
+- ZH: 同一诉请块内，后续以 `tellask* function call` 开头的行会被并入诉请头（不会开启新的诉请）。
 
-- EN: Put structured directives such as `!tellaskSession <slug>` in the headline.
-- ZH: 结构化指令（例如 `!tellaskSession <slug>`）必须放在诉请头中。
+- EN: Put structured directives such as `sessionSlug` in the headline.
+- ZH: 结构化指令（例如 `sessionSlug`）必须放在诉请头中。
 
 #### Tellask body（诉请正文）
 
-- EN: Lines in the Tellask block that start with `!?` but do not start with `!?@`. They carry the request context (steps, constraints, acceptance criteria, etc.).
-- ZH: 诉请块内以 `!?` 开头但不以 `!?@` 开头的行，用于承载诉请正文（步骤、上下文、约束、验收标准等）。
+- EN: The `tellaskContent` argument in tellask-special function calls. It carries request context (steps, constraints, acceptance criteria, etc.).
+- ZH: tellask-special 函数调用里的 `tellaskContent` 参数，用于承载诉请正文（步骤、上下文、约束、验收标准等）。
 
 ### 3 种诉请形态（Three Tellask Modes）
 
@@ -117,13 +117,13 @@
 - EN (what “Back” means): “Back” refers to routing back to the tellasker dialog; it does **not** imply hierarchy/seniority.
 - ZH（Back 的含义）: “Back” 指回到诉请者对话，**不暗示上下级**。
 
-- EN (typical carrier): `!?@tellasker ...` (only available inside a sideline dialog)
-- ZH（典型载体）: `!?@tellasker ...`（只在你处于支线对话语境时可用）
+- EN (typical carrier): `tellaskBack({ tellaskContent: "..." }) ...` (only available inside a sideline dialog)
+- ZH（典型载体）: `tellaskBack({ tellaskContent: "..." }) ...`（只在你处于支线对话语境时可用）
 
 Example / 示例（概念）:
 
-- EN: `!?@tellasker I need you to confirm the file extensions: only .md, or also .txt/.rst?`
-- ZH: `!?@tellasker 我需要你确认要扫描的文件扩展名：只包含 .md 还是也包含 .txt/.rst？`
+- EN: `tellaskBack({ tellaskContent: "..." }) I need you to confirm the file extensions: only .md, or also .txt/.rst?`
+- ZH: `tellaskBack({ tellaskContent: "..." }) 我需要你确认要扫描的文件扩展名：只包含 .md 还是也包含 .txt/.rst？`
 
 #### 2) Tellask Session（长线诉请）
 
@@ -135,8 +135,8 @@ Example / 示例（概念）:
 
 ##### 会话 Slug（Session Slug）
 
-- EN (directive; headline only): `!tellaskSession <slug>`
-- ZH（指令；仅 headline）: `!tellaskSession <slug>`
+- EN (directive; headline only): `sessionSlug`
+- ZH（指令；仅 headline）: `sessionSlug`
 
 - EN (parameter name concept): `tellaskSession` (parameter names are English-only; not i18n'd)
 - ZH（参数名概念）: `tellaskSession`（参数名只用英文，不做 i18n）
@@ -144,8 +144,8 @@ Example / 示例（概念）:
 - EN (slug format): short, stable, human-readable (e.g. `ws-schema-v2`, `tooling-read-file-ux`).
 - ZH（slug 格式）: 简短、稳定、可读（例如 `ws-schema-v2`、`tooling-read-file-ux`）。
 
-- EN (placement rule): Put `!tellaskSession` in the Tellask headline; do not put it on a second line (it would become body text).
-- ZH（位置规则）: `!tellaskSession` 必须写在诉请 headline 中；不要放到第二行（否则会进入 body 变成普通文本）。
+- EN (placement rule): Put `sessionSlug` in the Tellask headline; do not put it on a second line (it would become body text).
+- ZH（位置规则）: `sessionSlug` 必须写在诉请 headline 中；不要放到第二行（否则会进入 body 变成普通文本）。
 
 ##### 多人会话（Multi-Party Sessions）
 
@@ -161,11 +161,11 @@ Example / 示例（概念）:
 Example / 示例（概念）:
 
 ```plain-text
-!?@server !tellaskSession ws-schema-v2
+tellask({ targetAgentId: "server", sessionSlug: "ws-schema-v2", tellaskContent: "..." })
 !?Please confirm the WS packet schema versioning strategy and point to code anchors.
 !?请确认 WS packet schema 的版本化策略，并指出相关代码锚点。
 
-!?@webui !tellaskSession ws-schema-v2
+tellask({ targetAgentId: "webui", sessionSlug: "ws-schema-v2", tellaskContent: "..." })
 !?Explain which missing fields cause UX degradation along the current WebUI subscribe/render path.
 !?按当前 WebUI 订阅/渲染路径说明：哪些字段缺失会导致 UX 退化。
 ```
@@ -181,18 +181,18 @@ Example / 示例（概念）:
 - EN (key property): “Fresh/one-shot” is not only “new context”; it also means **no continuation semantics** — later Tellasks are not expected to resume the same session context.
 - ZH（关键性质）: “Fresh/一次性”不仅表示“新开上下文”，更表示：**没有后续续话语义** —— 后续诉请不应被期待能自动续接本次一次性诉请的上下文。
 
-- EN (practical guidance): If you need a follow-up after a Fresh Tellask, treat it as a new request and restate necessary context; if you need iterative follow-ups, use `Tellask Session` with `!tellaskSession <slug>`.
-- ZH（实践建议）: 如果你在一次性诉请后还要追问，应当把追问当作全新请求并补齐必要上下文；如果你需要可迭代的追问/推进，请使用 `Tellask Session` 并提供 `!tellaskSession <slug>`。
+- EN (practical guidance): If you need a follow-up after a Fresh Tellask, treat it as a new request and restate necessary context; if you need iterative follow-ups, use `Tellask Session` with `sessionSlug`.
+- ZH（实践建议）: 如果你在一次性诉请后还要追问，应当把追问当作全新请求并补齐必要上下文；如果你需要可迭代的追问/推进，请使用 `Tellask Session` 并提供 `sessionSlug`。
 
 Example / 示例（概念）:
 
-- EN: `!?@<shell-specialist> Please run a single build and paste the failure output.`
-- ZH: `!?@<shell-specialist> 请运行一次构建并回贴失败信息。`
+- EN: `tellaskSessionless({ targetAgentId: "shell-specialist", tellaskContent: "Please run a single build and paste the failure output." })`
+- ZH: `tellaskSessionless({ targetAgentId: "shell-specialist", tellaskContent: "请运行一次构建并回贴失败信息。" })`
 
 ### 系统提示可复用的一句话（One-Sentence Summary for System Prompts）
 
-- EN: `TellaskBack` asks the tellasker dialog for clarification; `Tellask Session` uses `!tellaskSession <slug>` for resumable multi-turn work; `Fresh Tellask` is one-shot and non-resumable.
-- ZH: `TellaskBack` 回问诉请者澄清；`Tellask Session` 用 `!tellaskSession <slug>` 进行可续用多轮协作；`Fresh Tellask` 是一次性且不可恢复。
+- EN: `TellaskBack` asks the tellasker dialog for clarification; `Tellask Session` uses `sessionSlug` for resumable multi-turn work; `Fresh Tellask` is one-shot and non-resumable.
+- ZH: `TellaskBack` 回问诉请者澄清；`Tellask Session` 用 `sessionSlug` 进行可续用多轮协作；`Fresh Tellask` 是一次性且不可恢复。
 
 ### 为何保留 `!` 前缀？（Why keep the `!` prefix?）
 
@@ -260,8 +260,8 @@ Example / 示例（概念）:
 
 ### Q4H (Question for Human) （向人类的诉请）
 
-- EN: A mechanism for raising questions to humans, initiated via `!?@human`, which suspends dialog progression until the human responds. **Always use "Q4H" (capital Q, numeral 4, capital H); never use "Q-for-H", "QforH", or "4-hour".**
-- ZH: 一种通过 `!?@human` 向人类提问的机制，暂停对话进度直到人类响应。**统一使用"Q4H"（大写 Q、数字 4、大写 H）；禁止使用"Q-for-H"、"QforH"、"每四小时"等变体。**
+- EN: A mechanism for raising questions to humans, initiated via `!?askHuman()`, which suspends dialog progression until the human responds. **Always use "Q4H" (capital Q, numeral 4, capital H); never use "Q-for-H", "QforH", or "4-hour".**
+- ZH: 一种通过 `!?askHuman()` 向人类提问的机制，暂停对话进度直到人类响应。**统一使用"Q4H"（大写 Q、数字 4、大写 H）；禁止使用"Q-for-H"、"QforH"、"每四小时"等变体。**
 
 ### Fresh Boots Reasoning（扪心自问）
 
@@ -402,16 +402,16 @@ Example / 示例（概念）:
 - EN: The implementation may still use the internal labels **Type A/B/C** to classify teammate-tellask patterns.
 - ZH: 实现层仍可能使用 **Type A/B/C** 作为队友诉请形态的内部分类。
 
-- EN: Type A: TellaskBack call (a subdialog asking back to its tellasker dialog); primary syntax `!?@tellasker` (NO `!tellaskSession`).
-- ZH: Type A：回问诉请（子对话回问其诉请者对话）；主语法 `!?@tellasker`（不带 `!tellaskSession`）。
+- EN: Type A: TellaskBack call (a subdialog asking back to its tellasker dialog); primary syntax `tellaskBack({ tellaskContent: "..." })` (NO `sessionSlug`).
+- ZH: Type A：回问诉请（子对话回问其诉请者对话）；主语法 `tellaskBack({ tellaskContent: "..." })`（不带 `sessionSlug`）。
 
-- EN: Type B: registered subdialog call (resumable) keyed by `agentId!tellaskSession`.
-- ZH: Type B：registered subdialog call（可恢复），用 `agentId!tellaskSession` 作为 registry key。
+- EN: Type B: registered subdialog call (resumable) keyed by `agentIdsessionSlug`.
+- ZH: Type B：registered subdialog call（可恢复），用 `agentIdsessionSlug` 作为 registry key。
 
 - EN: Type C: transient subdialog call (one-shot), not registered.
 - ZH: Type C：transient subdialog call（一次性），不注册到 registry。
 
-### `!tellaskSession` / 会话 Slug 指令
+### `sessionSlug` / 会话 Slug 指令
 
-- EN: Resumable registered subdialogs use `!tellaskSession <slug>` in the Tellask headline.
-- ZH: 可恢复的注册子对话使用 Tellask headline 内的 `!tellaskSession <slug>`。
+- EN: Resumable registered subdialogs use `sessionSlug` in the Tellask headline.
+- ZH: 可恢复的注册子对话使用 Tellask headline 内的 `sessionSlug`。

package/dist/docs/fbr-implementation.md

@@ -27,7 +27,7 @@ For FBR policy, runtime:
 - uses `buildFbrSystemPrompt(...)` (no tool instructions)
 - injects a separate `buildNoToolsNotice(...)`
 - forces `effectiveAgentTools = []`
-- forces `tellaskPolicy = tellasker_only`
+- forces `tellaskPolicy = deny_all`
 - forces `allowFunctionCalls = false`
 - applies `fbr_model_params` override when configured
 
@@ -39,7 +39,7 @@ For FBR policy, runtime:
 
 Both streaming and non-streaming paths call `resolveDrivePolicyViolationKind(...)` to detect:
 
-- disallowed tellasks (FBR allows only `@tellasker`)
+- disallowed tellask-special function calls (FBR technical mode disallows all function calls)
 - disallowed tool/function calls (FBR disallows all)
 
 On violation, runtime emits `formatDomindsNoteFbrToollessViolation(...)` consistently.
@@ -51,7 +51,7 @@ Before generation, runtime runs `validateDrivePolicyInvariants(...)` and fail-fa
 - system prompt must exactly equal `buildFbrSystemPrompt(...)`
 - `effectiveAgentTools` must be empty
 - `allowFunctionCalls` must be `false`
-- `tellaskPolicy` must be `tellasker_only`
+- `tellaskPolicy` must be `deny_all`
 - `prependedContextMessages` must contain exactly one `buildNoToolsNotice(...)`
 
 If any check fails, runtime throws `FBR policy isolation violation`, preventing global tool-manual/tool-prompt paths from leaking into FBR.
@@ -74,4 +74,4 @@ If any check fails, runtime throws `FBR policy isolation violation`, preventing
 - FBR system prompt has no tool instructions.
 - Tool-related wording appears only in the separate `buildNoToolsNotice(...)`.
 - FBR/non-FBR context assembly flow is structurally identical; only policy fields differ.
-- Any FBR tool/function call or non-`@tellasker` tellask is hard-rejected with explicit feedback.
+- Any FBR tool/function call or tellask-special function call is hard-rejected with explicit feedback.

package/dist/docs/fbr-implementation.zh.md

@@ -27,7 +27,7 @@
 - 切换到 `buildFbrSystemPrompt(...)`（不含工具说明）
 - 注入单独 `buildNoToolsNotice(...)`
 - 强制 `effectiveAgentTools = []`
-- 强制 `tellaskPolicy = tellasker_only`
+- 强制 `tellaskPolicy = deny_all`
 - 强制 `allowFunctionCalls = false`
 - 在需要时对成员应用 `fbr_model_params` 覆盖
 
@@ -39,7 +39,7 @@
 
 流式与非流式两条路径都调用 `resolveDrivePolicyViolationKind(...)`：
 
-- 违规 tellask（FBR 仅允许 `@tellasker`）
+- 违规 tellask-special 函数调用（FBR 技术模式下禁止全部函数调用）
 - 违规 function/tool call（FBR 禁止）
 
 一旦违规，统一产出 `formatDomindsNoteFbrToollessViolation(...)`，保持用户反馈与日志语义一致。
@@ -51,7 +51,7 @@
 - system prompt 必须严格等于 `buildFbrSystemPrompt(...)`
 - `effectiveAgentTools` 必须为空
 - `allowFunctionCalls` 必须为 `false`
-- `tellaskPolicy` 必须为 `tellasker_only`
+- `tellaskPolicy` 必须为 `deny_all`
 - `prependedContextMessages` 必须且仅能包含一条 `buildNoToolsNotice(...)`
 
 若任一条件不满足，运行时直接抛出 `FBR policy isolation violation`，防止全局工具手册/工具提示路径回流污染 FBR。
@@ -74,4 +74,4 @@
 - FBR system prompt 不包含工具说明。
 - “无工具”文案仅来自独立 `buildNoToolsNotice(...)`。
 - FBR 与非 FBR 的上下文装配主流程一致，差异只来自策略字段。
-- FBR 中任意 tool/function call 或非 `@tellasker` tellask 都被运行时硬拒绝并给出明确回执。
+- FBR 中任意 tool/function call 或 tellask-special 函数调用都被运行时硬拒绝并给出明确回执。

package/dist/docs/fbr.md

@@ -9,8 +9,8 @@ Chinese version: [中文版](./fbr.zh.md)
 **Fresh Boots Reasoning (FBR)** is a Dominds mechanism for “reasoning again from a clean slate” on a bounded sub-problem,
 then reporting back to the mainline dialog.
 
-In Dominds, FBR is triggered via Tellask syntax. `@self` is only the entry-point syntax (“tellask the current dialog
-itself”); the mechanism is the runtime-enforced contract applied to the spawned sideline dialog(s).
+In Dominds, FBR is triggered by the dedicated function tool `freshBootsReasoning({ tellaskContent: "..." })`.
+The mechanism is the runtime-enforced contract applied to the spawned sideline dialog(s).
 
 ## 2. Design principles and tradeoffs
 
@@ -24,7 +24,7 @@ predictable, FBR sideline dialogs must be:
 
 ### 2.2 No silent failure
 
-If FBR is disabled by configuration (e.g. `fbr-effort: 0`), the runtime MUST reject `!?@self` loudly and clearly. A
+If FBR is disabled by configuration (e.g. `fbr-effort: 0`), the runtime MUST reject `freshBootsReasoning({ tellaskContent: "..." })` loudly and clearly. A
 silent ignore is worse than an error.
 
 ### 2.3 Many-shot reasoning, not “multi-agent collaboration”
@@ -32,28 +32,25 @@ silent ignore is worse than an error.
 `fbr-effort` is for producing multiple _independent_ reasoning samples in parallel. The mainline dialog is responsible
 for synthesis; FBR sidelines do not coordinate with each other.
 
-## 3. User syntax (`@self` is just the entry point)
+## 3. User syntax
 
 ### 3.1 Trigger forms
 
-Both forms below trigger FBR:
+Use the dedicated FBR form:
 
-- **Default (transient)**: `!?@self`
-- **Rare (resumable)**: `!?@self !tellaskSession <tellaskSession>`
+- `freshBootsReasoning({ tellaskContent: "..." })`
 
 Notes:
 
-- `@self` means “tellask the current dialog itself (same agentId)”. It is an explicit self-tellask marker to avoid
-  accidental self-calls caused by echoing/quoting `@teammate` headlines.
-- Resumability comes from `!tellaskSession`, not from `@self`.
+- FBR does not use `targetAgentId`, `sessionSlug`, or `mentionList`.
+- `tellaskContent` is the authoritative task context for the FBR sideline.
 
 ### 3.2 Scope
 
-This document specifies the FBR mechanism and its `!?@self` contract. General teammate Tellasks (`!?@<teammate>`) follow
+This document specifies the FBR mechanism and its `freshBootsReasoning({ tellaskContent: "..." })` contract. General teammate Tellasks (`tellaskSessionless({ targetAgentId: "<teammate>", tellaskContent: "..." })`) follow
 the taxonomy and capability model in [`dialog-system.md`](./dialog-system.md).
 
-If you want a “fresh” sideline dialog that still has tools, do not use `@self`. Use an explicit teammate identity that
-is granted the needed toolsets (or use the general `!?@<agentId>` mechanism intentionally).
+If you want a “fresh” sideline dialog that still has tools, use an explicit teammate identity via the general teammate Tellask flow.
 
 ## 4. Runtime contract (normative)
 
@@ -61,20 +58,17 @@ This section uses MUST / MUST NOT / SHOULD / MAY for requirements.
 
 ### 4.1 Isolation and context
 
-When driving an FBR sideline dialog created by `!?@self`, runtime MUST enforce:
+When driving an FBR sideline dialog created by `freshBootsReasoning({ tellaskContent: "..." })`, runtime MUST enforce:
 
 - **No dependency on tellasker dialog history**
   - the tellaskee MUST NOT assume access to the tellasker’s mainline/sideline history
   - the tellaskee MUST treat the tellask body as the primary, authoritative task context
-  - for the resumable `!tellaskSession` form, the tellaskee MAY use the sideline’s _own_ `tellaskSession` history as
-    explicit context
 - **No tool-based context fetch**
   - no reading files / running commands / browsing
   - no accessing Memory or rtws (runtime workspace) state
 
 Intuition: “fresh boots” means “fresh relative to the caller thread”, not “ignores baseline system rules”. Runtime may
-still inject baseline policy/safety/formatting context, but the tellask body remains the authority (except when using
-the sideline’s own resumable history).
+still inject baseline policy/safety/formatting context, but the tellask body remains the authority.
 
 ### 4.2 Tool-less (prompt + technical enforcement)
 
@@ -90,7 +84,7 @@ The FBR system prompt MUST communicate (wording may vary, meaning must hold):
 - this is an FBR sideline dialog; the tellask body is the primary context
 - do not assume access to tellasker dialog history
 - if critical context is missing, list what is missing and why it blocks reasoning
-- `!?@tellasker` is allowed only when you must clarify critical missing context; otherwise do not emit any tellasks
+- `tellaskBack({ tellaskContent: "..." })` is allowed only when you must clarify critical missing context; otherwise do not emit any tellasks
 
 And: the **system prompt body MUST NOT include tool instructions** (no tool lists, allowlists, example commands, “how to
 use tools”, etc.).
@@ -113,16 +107,16 @@ Under no circumstances should the FBR sideline dialog see any tool definitions.
 
 #### 4.2.3 The LLM request MUST be “zero tools”
 
-The LLM request for an `@self` FBR sideline dialog MUST have **zero tools available**:
+The LLM request for an FBR sideline dialog (`freshBootsReasoning`) MUST have **zero tools available**:
 
 - the request payload must not include tool/function definitions (effective tool list must be empty)
 - provider tool-calling / function-calling modes must not be enabled
 
 If the model attempts a tool/function call anyway, runtime MUST hard-reject it (see 4.5).
 
-### 4.3 Tellask restriction: only `!?@tellasker`
+### 4.3 Tellask restriction: only `tellaskBack({ tellaskContent: "..." })`
 
-FBR sideline dialogs MUST NOT issue teammate Tellasks (including `!?@human`). The only exception is `!?@tellasker`:
+FBR sideline dialogs MUST NOT issue teammate Tellasks (including `askHuman({ tellaskContent: "..." })`). The only exception is `tellaskBack({ tellaskContent: "..." })`:
 
 - sideline-only TellaskBack to the upstream tellasker dialog
 - allowed only when critical context must be clarified
@@ -133,14 +127,14 @@ FBR sideline dialogs MUST NOT issue teammate Tellasks (including `!?@human`). Th
 An FBR sideline dialog should produce a compact artifact that is easy for the tellasker to integrate. Suggested shape:
 
 1. **Conclusion**
-2. **Reasoning** (grounded in the tellask body; plus resumable history when applicable)
+2. **Reasoning** (grounded in the tellask body)
 3. **Assumptions** (explicitly sourced: body vs session history)
 4. **Unknowns / missing context**
 5. **Next steps for mainline** (where tools/teammates may exist)
 
 ### 4.5 Violations and errors (loud + debuggable)
 
-- Any disallowed tellask (non-`@tellasker`) or any tool/function call attempt inside FBR MUST be treated as a hard
+- Any disallowed tellask (anything other than `tellaskBack({ tellaskContent: "..." })`) or any tool/function call attempt inside FBR MUST be treated as a hard
   violation.
 - The runtime MUST return a clear, user-visible error, and MUST log/emit a debuggable reason string (no silent swallow).
 
@@ -150,25 +144,22 @@ An FBR sideline dialog should produce a compact artifact that is easy for the te
 
 - Type: integer
 - Default: `3`
-- `0`: disable `!?@self` FBR for that member (runtime MUST reject `!?@self` clearly)
-- `1..100`: spawn N FBR sideline dialogs per `!?@self`
+- `0`: disable `freshBootsReasoning({ tellaskContent: "..." })` FBR for that member (runtime MUST reject `freshBootsReasoning({ tellaskContent: "..." })` clearly)
+- `1..100`: spawn N FBR sideline dialogs per `freshBootsReasoning({ tellaskContent: "..." })`
 - `> 100` / non-integer / negative: validation error (reject; no clamping)
 
 When `fbr-effort = N`:
 
-- runtime expands a single `!?@self` into **N parallel tool-less FBR sideline dialogs**
+- runtime expands a single `freshBootsReasoning({ tellaskContent: "..." })` into **N parallel tool-less FBR sideline dialogs**
 - each sideline receives the same tellask body and the same tool-less constraints
 - mainline receives all N responses; **ordering must not be relied on** (completion order is fine)
 
-For resumable `!?@self !tellaskSession <tellaskSession>` with `N > 1`, runtime MUST derive **distinct** session keys so
-parallel sidelines do not share history (recommended: `<tellaskSession>.fbr-<i>`).
-
 ## 6. FBR-only model overrides: `fbr_model_params`
 
 `fbr_model_params` overrides model params **only when driving FBR sideline dialogs**:
 
 - Schema: identical to `model_params` (documented by `model_param_options` in `dominds/main/llm/defaults.yaml`)
-- Scope: `!?@self` only
+- Scope: `freshBootsReasoning({ tellaskContent: "..." })` only
 - Merge: recommended deep-merge on top of the member’s effective `model_params`
 - `max_tokens` may be configured as top-level `max_tokens` or `general.max_tokens` (pick one; do not set both)
 
@@ -179,38 +170,25 @@ parallel sidelines do not share history (recommended: `<tellaskSession>.fbr-<i>`
 Bad (depends on external context/tools):
 
 ```text
-!?@self
-Find the bug and fix it.
+freshBootsReasoning({ tellaskContent: "Find the bug and fix it." })
 ```
 
 Good (puts the actual context into the body):
 
 ```text
-!?@self
-You are doing tool-less FBR. Use ONLY the text below.
-
-Goal: identify the most likely root cause and propose 2–3 viable fixes.
-
-Observed:
-- Clicking “Run” sometimes freezes the UI for ~10s.
-
-Constraint:
-- We cannot change the backend protocol.
-
-Evidence:
-<paste relevant logs / code / stack trace here>
+freshBootsReasoning({ tellaskContent: "You are doing tool-less FBR. Use ONLY the text below.\n\nGoal: identify the most likely root cause and propose 2–3 viable fixes.\n\nObserved:\n- Clicking “Run” sometimes freezes the UI for ~10s.\n\nConstraint:\n- We cannot change the backend protocol.\n\nEvidence:\n<paste relevant logs / code / stack trace here>" })
 ```
 
 ### 7.2 `.minds/team.yaml`
 
 ```yaml
 member_defaults:
-  # Spawn 3 tool-less FBR sideline dialogs per `!?@self` by default.
+  # Spawn 3 tool-less FBR sideline dialogs per `freshBootsReasoning({ tellaskContent: "..." })` by default.
   fbr-effort: 3
 
 members:
   ux:
-    # Spawn 5 independent reasoning samples per `!?@self`.
+    # Spawn 5 independent reasoning samples per `freshBootsReasoning({ tellaskContent: "..." })`.
     fbr-effort: 5
 
     # Make FBR more exploratory without changing mainline behavior.
@@ -224,14 +202,14 @@ members:
 
 ## 8. Relationship to general sideline dialogs
 
-- `!?@self` is a special case: tool-less, body-first, tellask-restricted, optionally fanned out via `fbr-effort`.
-- General `!?@<teammate>` sidelines remain fully capable (tools/toolsets as configured).
-- If you need “same persona + tools” in a sideline, do not use `@self`; use an explicit teammate identity.
+- `freshBootsReasoning({ tellaskContent: "..." })` is a special case: tool-less, body-first, tellask-restricted, optionally fanned out via `fbr-effort`.
+- General `tellaskSessionless({ targetAgentId: "<teammate>", tellaskContent: "..." })` sidelines remain fully capable (tools/toolsets as configured).
+- If you need “same persona + tools” in a sideline, use an explicit teammate identity (`tellask` / `tellaskSessionless`).
 
 ## 9. Acceptance checklist
 
-- `!?@self` triggers tool-less FBR; the LLM request is technically “zero tools”.
+- `freshBootsReasoning({ tellaskContent: "..." })` triggers tool-less FBR; the LLM request is technically “zero tools”.
 - The system prompt body contains no tool instructions; tool-related wording comes only from the separate fixed notice.
-- FBR sidelines cannot issue teammate Tellasks; only `!?@tellasker` is allowed when necessary.
+- FBR sidelines cannot issue teammate Tellasks; only `tellaskBack({ tellaskContent: "..." })` is allowed when necessary.
 - `fbr-effort` defaults to `3`, accepts `0..100`, rejects invalid values, and fails loudly when disabled.
 - `fbr_model_params` applies only to FBR and follows the same schema/merge intent as `model_params`.