@danceiny/gotry 0.0.1-rc.18 → 0.0.1-rc.20

package/README.md

@@ -13,7 +13,7 @@ GoTry is an AI travel agent for **"departure to next departure."** You tell it w
 [![Node](https://img.shields.io/badge/node-%E2%89%A5%2022.15-blue)](https://www.npmjs.com/package/@danceiny/gotry)
 [![Docs](https://img.shields.io/badge/docs-architecture.md-blue)](docs/architecture.md)
 
-**[What GoTry Does](#what-gotry-does)** · **[How It Works](#how-it-works)** · **[Tools](#tools)** · **[Demo](#demo)** · **[Quick Start](#quick-start)** · **[Consent and Privacy](#consent-and-privacy)** · **[Trustworthy by Construction](#trustworthy-by-construction)** · **[Project Status](#project-status)** · **[Roadmap](#roadmap)** · **[For AI Agents](#for-ai-agents)** · **[Documentation](#documentation)** · **[License](#license)**
+**[What GoTry Does](#what-gotry-does)** · **[How It Works](#how-it-works)** · **[Tools](#tools)** · **[Demo](#demo)** · **[Benchmark](#how-mainstream-ai-answers-the-same-trip)** · **[Quick Start](#quick-start)** · **[Consent and Privacy](#consent-and-privacy)** · **[Trustworthy by Construction](#trustworthy-by-construction)** · **[Project Status](#project-status)** · **[Roadmap](#roadmap)** · **[For AI Agents](#for-ai-agents)** · **[Documentation](#documentation)** · **[License](#license)**
 
 > **Tip for newcomers:** one command is enough to feel the difference — `npx @danceiny/gotry web`, open `http://127.0.0.1:3080`, and say *"I want three relaxing days in Dali."* The agent interviews you first; then the solver, not the model, decides what is feasible. Full walkthrough: [`docs/user-guide.md`](docs/user-guide.md).
 
@@ -69,12 +69,12 @@ Architecture, five layers:
 
 ## Tools
 
-21 tools in six groups:
+22 tools in six groups:
 
 | Group | Tool | What it does |
 |---|---|---|
-| **Realtime retrieval (OTA/official, read-only)** | `gotry_flyai_search` | Live flight/train/hotel quotes via the Fliggy official channel (masked hotel prices upstream; real prices on the jumpUrl page) |
-| | `gotry_session_search` | Ctrip flights on the **user's own logged-in Chrome session** (consent-gated, physically read-only) |
+| **Realtime retrieval (OTA/official, read-only)** | `gotry_flyai_search` | Live flight/train/hotel quotes via the Fliggy official channel (masked hotel prices upstream; real prices on the jumpUrl page; exhausted anonymous trial quota degrades as `needs-setup` with key guidance, never silent retries) |
+| | `gotry_session_search` | Ctrip flights **and hotels** + 12306 trains on the **user's own Chrome session** (consent-gated, physically read-only; hotels = `kind:"hotel"` + optional `cityId`, real logged-in prices; trains = `kind:"train"`, public query face — codes/times/seat availability, no prices in the list API) |
 | | `gotry_session_login` | Login bootstrap: auto-detects existing login first; otherwise opens the login entry in the user's Chrome (**zero terminal**) |
 | | `gotry_weather_check` | Open-Meteo forecast ≤16 d + historical climate baseline |
 | | `gotry_flight_verify` | OpenSky ADS-B live flight observation (three-valued) |
@@ -88,6 +88,9 @@ Architecture, five layers:
 | **Artifacts** | `gotry_artifacts_list` / `gotry_artifacts_read` | Discover & view generated artifacts (async deliverables + working-dir markdown) as a line-numbered file view (read-only) |
 | **Factuality gate** | `gotry_fact_gate` | Pre-delivery gate for itinerary artifacts — see [fact gate](#how-it-works) above |
 | **General external** | `gotry_web_search` · `gotry_video_subtitle` · `gotry_github_search` · `gotry_agent_reach` | web / subtitles / GitHub / all-channel external info (via Agent-Reach) |
+| **Self-check** | `gotry_doctor` | Read-only health check of optional dependencies (extension / Agent-Reach .venv / hbcli / FlyAI key **+ recent trial-quota exhaustion time** / dsh-calendar mount state / sidebar / dsh-map-tools & dsh-tool-ask-user presence) with exact repair guidance; installs only ever happen via the user running `npx gotry doctor --fix`. LLM keys are the dsh host's business — deliberately out of scope. Report lands in `gotry-state/doctor-report.md` (sidebar-workbench previewable) |
+
+> **Channel routing**: retrieval tools stay flat (no hidden dispatch); the persona routing card and the `routing` suggestions attached to failed search results are **generated from one channel registry** (official API > user session > web fallback, filtered by per-session channel health). When a channel exhausts its quota the result says so and names the next channel — retry-blindness is a contract violation, not a prompt hope.
 
 ## Demo
 
@@ -111,6 +114,21 @@ Engine verdict:
 
 > Tag guide: `[skeleton:openflights]` means "this route can be flown" was verified against the public route database; `[realtime:...]` marks data pulled live seconds ago; `[static-pack:estimate]` flags an off-season estimate — **verify before booking**. Tags are attached by the render layer, never by the model.
 
+## How Mainstream AI Answers the Same Trip
+
+GoTry's product persona is calibrated against evidence, not taste: the same real three-week multi-country workation prompt — with planted traps (no year given, a vague "some seaside town called Wan-xx", an ambiguity the user already resolved) — is fed verbatim to mainstream assistants, their answers archived word-for-word, and scored against a ground-truth rubric. Transcripts, rubric, and the contract feedback: [`docs/persona-bench/`](docs/evaluation/persona-bench/).
+
+| Dimension | Generic chat assistant (Kimi, 13 real turns) | OTA agent (Fliggy open platform, single turn) | GoTry contract |
+|---|---|---|---|
+| Calendar grounding | ✗ 2025 calendar; three user corrections, three apology-refits | ✗ derived weekdays land on the 2025 calendar — contradicting its own answer on the same page | (2)(8)(9) + time-anchor card |
+| Constraint interview | ✗ zero questions; both load-bearing constraints surfaced by the user at turn 6 | △ asks sales qualifiers (budget / star level / sea view); zero must-asks | (1)(10) |
+| Feasibility & time accounting | ✗ density illusion, caught by the user | ✗ HK errands + same-day flight with no time budget; an "8h" flight contradicting its own arrival time | (4) + door-to-door true cost |
+| Fact provenance | △ destination research holds up | ✗ sells a defunct airline (retired 2020); every price unsourced | (3)(7)(13)(20) |
+| Structure completeness | △ decent comparison table only at turn 13 | ✓✓ full skeleton in one turn — completeness is table stakes | verified completeness (fact gate) |
+| Persona in one line | erudite but stateless chatter — the user ends up doing four jobs | a flawless-brochure OTA clerk — every section ends in a price table | trusted travel engineer: interview first, the solver decides, infeasible says infeasible |
+
+**Single best finding: two unrelated products derived their weekdays from the 2025 calendar.** Calendar grounding has to be a product mechanism (anchor card, assert once, never recompute) — not model luck. Cautionary deep-dive: [`docs/research/kimi-postmortem.md`](docs/research/kimi-postmortem.md).
+
 ## Quick Start
 
 ### npm (recommended)
@@ -125,11 +143,16 @@ npx @danceiny/gotry web
 |---|---|---|
 | Web chat (recommended) | `npx @danceiny/gotry web` | multi-turn planning with visualized reasoning → `:3080` |
 | Headless one-shot | `npx @danceiny/gotry "Two recovery days from Shenzhen, budget 3000"` | scripts / CI / targeted debugging → stdout |
+| Dependency doctor | `npx @danceiny/gotry doctor` (`--fix` to repair) | optional channels misbehaving: checks extension / Agent-Reach / hbcli / FlyAI key / sidebar / dsh-calendar mount / dsh-map-tools & dsh-tool-ask-user presence, prints exact repair guidance, writes `gotry-state/doctor-report.md` (previewable in the sidebar workbench) |
 
 Requires Node ≥ 22.15. LLM credentials are managed by your dsh host UI — gotry itself never asks for or echoes them. OpenAI-compatible endpoints (MiniMax / relays / self-hosted gateways) are handled by the dsh model configuration. First cold start takes 6–15 s; if port `:3080` is taken, free it first; unexpected exits leave evidence in `gotry-state/incidents.jsonl` (nothing silent).
 
 > **Cost accounting** — `ts/data/llm-price-table.json` (schema `gotry_llm_price_table_v2`) is the single source of truth for nightly run cost. Adding a model or switching relays = a PR against this file (peak-conservative upper bounds only); unknown models **fail closed** — no guessed prices. Drift monitor: `npx tsx ts/scripts/price-drift-watch.ts` (offline baseline diff; `--fetch` for live official pages). It never auto-applies changes.
 
+> **Quality metrics (repo-side)** — `npx tsx ts/scripts/build-metrics-report.ts [--state-root <root>] [--out report.md] [--days 7]` aggregates the persisted sidecars (fact-gate verdict distribution & blocked rate, channel down/cooldown, incidents, bridge latency vs the 500 ms re-audit budget, ledger / doctor-report presence) into one read-only markdown report. No new dependencies, zero LLM; the state root is never written (only `--out` produces a file, outside the state root).
+
+> **Channel probe tick (out-of-band health)** — `npx tsx ts/scripts/channel-probe.ts --state-root <root>` runs one read-only probe round over headless-probeable channels (hbcli whoami / open-meteo / opensky; session surfaces are skipped, FlyAI is not probed by default to preserve the shared anonymous quota) and appends `down` / `'ok'` recovery events to `channel-health.jsonl` — routing advice and doctor pick them up with zero changes. Driven by cron/loopx; no resident process. Wish recall consumes the same facts: a wish whose `conditions.channels` names a currently-down channel is vetoed for that recall. Remote (world2agent) callback stays gated on D-31.
+
 ### Developer source install
 
 ```bash
@@ -165,23 +188,24 @@ One-time prerequisite: the [GoTry Session Bridge](https://chromewebstore.google.
 
 ## Project Status
 
-Current release: **v0.0.1-rc.17** (npm dist-tag `rc`; `latest` stays on rc.16 per the dsh-as-LLM-host install convention). Evaluation is at Phase 0 foundation — deterministic contracts, validators, and a cadence policy; no external benchmark scores, no spend, no uplift claims.
+Current release: **v0.0.1-rc.18** (npm `latest` and `rc` both point here; registry pull-verified 2026-09-03: clean install / bin / dist entry all pass). Evaluation is at Phase 0 foundation — deterministic contracts, validators, and a cadence policy; no external benchmark scores, no spend, no uplift claims.
 
 **Working today** (full-stack regression green; every item has deterministic tests):
 
 - **Z3 solving engine** — feasibility verdicts + door-to-door whole-cost; the historical concurrency race is fixed and regression-gated
-- **Realtime retrieval** — flights/trains/hotels (Fliggy official channel), destination/hotel catalogs, weather, live flight observation, route connectivity; realtime prices can overwrite solver prices (`GOTRY_REALTIME_PRICING=1`)
-- **Account-session search** — Ctrip flights on your logged-in Chrome; observed runs scored every landed hit 13/13 with zero write attempts, while non-hits stay explicit `miss` records — no live-availability claim beyond that
+- **Realtime retrieval** — flights/trains/hotels (Fliggy official channel), destination/hotel catalogs, weather, live flight observation, route connectivity; realtime prices can overwrite solver prices (`GOTRY_REALTIME_PRICING=1`); exhausted FlyAI anonymous trial quota is classified `needs-setup` with key guidance (no blind retries)
+- **Dependency doctor** — `npx gotry doctor` (CLI) / `gotry_doctor` (in-chat tool): read-only health check of optional dependencies (extension / Agent-Reach / hbcli / FlyAI key / sidebar / dsh-calendar mount / dsh-map-tools & dsh-tool-ask-user presence) with exact repair guidance; `--fix` installs; LLM keys stay with the dsh host
+- **Account-session search** — Ctrip flights **and hotels** + 12306 trains on your Chrome (hotels 2026-09-03: real logged-in prices via passive sniffing; trains 2026-09-03: public left-ticket query; interface surfaces calibrate with the first live session); observed runs scored every landed hit 13/13 with zero write attempts, while non-hits stay explicit `miss` records — no live-availability claim beyond that
 - **Extension install on demand** — `[GoTry Session Bridge](https://chromewebstore.google.com/detail/gotry-session-bridge/oeajpiccmonococjcegddlooeeohlbgd)` is offered as a clickable link in the dsh UI when an account-session tool first needs it (one-click install + auto-update); the gotry side never runs a setup wizard
 - **Memory & reachability** — motivation profile / wish pool / companions / travel timeline; English solve output via `GOTRY_LOCALE=en`
-- **Bounded agent tool loops** — soft convergence after 16 dispatches, structured `TOOL_BUDGET_EXHAUSTED` refusals beyond 18, exercised end-to-end through a packaged consumer install in CI
+- **Routed turn budgets** — every turn is classified (quick / sync / deep-planning) by a deterministic, zero-LLM router; time is the only budget and the deadline exit follows the task: quick and sync turns converge to an answer, deep-planning turns hand off to a persisted background ticket (`gotry_turn_handoff.v1`, ETA ≈1h) instead of dying mid-stream; the ticket is collected in the background by `scripts/turn-handoff-collect.ts` (idempotent, recursion-guarded child planner) and surfaces in-chat via the read-only `gotry_turn_handoff_list` tool; exercised end-to-end through a packaged consumer install in CI
 
 **Open limitations** (honest list):
 
 - **M3 Exit not closed** — engineering & distribution are ready, but real seed-user evidence (50–200 person cohort) has not been accumulated; automated tests prove contracts and formulas, not business pass
 - **Hotel session adapters** — Ctrip-hotel / Meituan logged-in surfaces await real login-state backfill; flights are done
 - **Interface language** — English covers the deterministic solve-output layer; the dsh host UI and dialogue surface belong to the host / calibration samples
-- **External benchmark generalization** — every frozen external run to date remains diagnostic-only (no score, no uplift claim); the round-by-round engineering ledger lives in [`docs/benchmark-environment-bridge.md`](docs/benchmark-environment-bridge.md)
+- **External benchmark generalization** — every frozen external run to date remains diagnostic-only (no score, no uplift claim). The bridge now combines the generic `tools/call/errors` recovery surface with per-tool typed descriptors, exact result/domain/failure conformance, and a latest-response terminal fence; the round-by-round engineering ledger lives in [`docs/evaluation/benchmark-environment-bridge.md`](docs/evaluation/benchmark-environment-bridge.md)
 - **Booking** — nothing bookable ships today; M5 opens only through WriteGate and the booking-saga FSM
 
 <details>
@@ -214,7 +238,7 @@ npx tsx scripts/evaluation-contract-tests.ts   # evaluation Phase 0 contracts (o
 npx tsx scripts/evaluation-cadence-tests.ts    # deterministic cadence policy/planner
 ```
 
-The suite covers golden engines, dialogue replay, cross-process async work-orders, plugin smoke, realtime bridges, process guards, i18n, memory domain, the Z3 concurrency gate, the fact gate, and a packaged-consumer tool-budget E2E, among others; the authoritative section list is whatever `scripts/run-all-tests.sh` enumerates. The live session benchmark (`npx tsx scripts/sf-live-benchmark.ts --golden=static`) is opt-in, requires your connected Chrome session, and never runs in CI.
+The suite covers golden engines, dialogue replay, cross-process async work-orders, plugin smoke, realtime bridges, process guards, i18n, memory domain, the Z3 concurrency gate, the fact gate, and a packaged-consumer turn-deadline E2E, among others; the authoritative section list is whatever `scripts/run-all-tests.sh` enumerates. The live session benchmark (`npx tsx scripts/sf-live-benchmark.ts --golden=static`) is opt-in, requires your connected Chrome session, and never runs in CI.
 
 ## Contributing
 
@@ -235,17 +259,19 @@ Program-level context: [`docs/gotry-master-outline.md`](docs/gotry-master-outlin
 
 | Document | Purpose |
 |---|---|
+| [`docs/README.md`](docs/README.md) | Docs conventions & full index (taxonomy, naming, lifecycle) |
 | [`docs/architecture.md`](docs/architecture.md) | System, ADRs, evolution, debt ledger (Chinese, authoritative) |
 | [`docs/gotry-master-outline.md`](docs/gotry-master-outline.md) | Program master outline & reuse matrix |
 | [`docs/gotry-product-design.md`](docs/gotry-product-design.md) | Product design: main loop, transparency, whole-cost model |
 | [`docs/roadmap.md`](docs/roadmap.md) | M0–M6 timeline & current position |
 | [`docs/user-guide.md`](docs/user-guide.md) | End-user guide |
 | [`docs/data-sources.md`](docs/data-sources.md) | Data sources & evidence-chain policy |
-| [`docs/extension-privacy.md`](docs/extension-privacy.md) | Session Bridge extension privacy |
-| [`docs/benchmark-environment-bridge.md`](docs/benchmark-environment-bridge.md) | External benchmark bridge — engineering ledger |
-| [`docs/evaluation-foundation.md`](docs/evaluation-foundation.md) | Evaluation Phase 0 foundation |
-| [`docs/booking-saga-fsm.md`](docs/booking-saga-fsm.md) | Booking saga FSM (the M5 seam vocabulary) |
-| [`docs/kimi-postmortem.md`](docs/kimi-postmortem.md) | A real AI-travel-planning failure postmortem (cautionary tale) |
+| [`docs/ops/extension-privacy.md`](docs/ops/extension-privacy.md) | Session Bridge extension privacy |
+| [`docs/evaluation/benchmark-environment-bridge.md`](docs/evaluation/benchmark-environment-bridge.md) | External benchmark bridge — engineering ledger |
+| [`docs/evaluation/evaluation-foundation.md`](docs/evaluation/evaluation-foundation.md) | Evaluation Phase 0 foundation |
+| [`docs/design/booking-saga-fsm.md`](docs/design/booking-saga-fsm.md) | Booking saga FSM (the M5 seam vocabulary) |
+| [`docs/research/kimi-postmortem.md`](docs/research/kimi-postmortem.md) | A real AI-travel-planning failure postmortem (cautionary tale) |
+| [`docs/evaluation/persona-bench/`](docs/evaluation/persona-bench/) | Agent-persona benchmark — same real-trip prompt answered by mainstream AIs: transcripts, scoring rubric, and the persona it shapes |
 | [`docs/release-notes.md`](docs/release-notes.md) | Release decisions per version (the "why") |
 | [`CHANGELOG.md`](CHANGELOG.md) | Machine-derived changelog (Keep a Changelog + Conventional Commits) |
 | [`docs/tokens.md`](docs/tokens.md) | npm 2FA / release mechanics |
@@ -258,4 +284,4 @@ Program-level context: [`docs/gotry-master-outline.md`](docs/gotry-master-outlin
 
 **Built with**: DeepSeek Harness 0.1.2-alpha.3 (root-pinned) · Cordis · Z3 (WASM) · loopx (pipx) · hotelbyte-cli · Agent-Reach v1.5.0 · OpenFlights · TypeScript
 
-**Version baseline: `v0.0.1-rc.16` (npm `latest`).** The authoritative verification gates for the current checkout are enumerated by `scripts/run-all-tests.sh`; release flow: `scripts/publish-npm.sh`.
+**Version baseline: `v0.0.1-rc.18` (npm `latest`).** The authoritative verification gates for the current checkout are enumerated by `scripts/run-all-tests.sh`; release flow: `scripts/publish-npm.sh`.

package/README.zh-CN.md

@@ -13,7 +13,7 @@
 [![Node](https://img.shields.io/badge/node-%E2%89%A5%2022.15-blue)](https://www.npmjs.com/package/@danceiny/gotry)
 [![Docs](https://img.shields.io/badge/docs-architecture.md-blue)](docs/architecture.md)
 
-**[它做什么](#它做什么)** · **[工作原理](#工作原理)** · **[工具](#工具)** · **[一段对话](#一段对话)** · **[快速开始](#快速开始)** · **[账号会话:授权与隐私](#账号会话授权与隐私)** · **[构造上可信](#构造上可信)** · **[状态与限制](#状态与限制)** · **[路线图](#路线图)** · **[给 AI Agent](#给-ai-agent)** · **[文档](#文档)** · **[License](#license)**
+**[它做什么](#它做什么)** · **[工作原理](#工作原理)** · **[工具](#工具)** · **[一段对话](#一段对话)** · **[同题横评](#与主流-ai-同题横评产品人格从哪来)** · **[快速开始](#快速开始)** · **[账号会话:授权与隐私](#账号会话授权与隐私)** · **[构造上可信](#构造上可信)** · **[状态与限制](#状态与限制)** · **[路线图](#路线图)** · **[给 AI Agent](#给-ai-agent)** · **[文档](#文档)** · **[License](#license)**
 
 > **新读者提示**:一条命令就能感受差别——`npx @danceiny/gotry web`,打开 `http://127.0.0.1:3080`,说一句「我想去大理躺三天」。Agent 会先访谈你;然后由求解器(而非模型)判定什么可行。完整走查见 [`docs/user-guide.md`](docs/user-guide.md)。
 
@@ -69,12 +69,12 @@ GoTry 把「想去哪」变成「能不能——怎么去、真实代价是多
 
 ## 工具
 
-六组共 21 个:
+六组共 22 个:
 
 | 组 | 工具 | 干什么 |
 |---|---|---|
-| **实时检索(OTA/官方只读)** | `gotry_flyai_search` | 机票/火车/酒店实时报价(飞猪官方通道;酒店价格为上游打码展示,真实价以 jumpUrl 为准) |
-| | `gotry_session_search` | 在**用户本人登录态**里查携程机票(授权后,只读) |
+| **实时检索(OTA/官方只读)** | `gotry_flyai_search` | 机票/火车/酒店实时报价(飞猪官方通道;酒店价格为上游打码展示,真实价以 jumpUrl 为准;匿名试用额度达限归类 `needs-setup` 并带配 key 指引,不盲重试) |
+| | `gotry_session_search` | 在**用户本人登录态**里查携程机票/酒店 + 12306 火车(kind=flight|hotel|train;授权后,只读;酒店为登录态真实价,火车为公开余票查询面) |
 | | `gotry_session_login` | 登录引导:自动检测已登录与否,未登录才在用户 Chrome 弹登录入口(**零终端**) |
 | | `gotry_weather_check` | Open-Meteo 预报≤16 天 + 历史气候基线 |
 | | `gotry_flight_verify` | OpenSky ADS-B 航班实时观测(三值) |
@@ -88,6 +88,9 @@ GoTry 把「想去哪」变成「能不能——怎么去、真实代价是多
 | **产物** | `gotry_artifacts_list` / `gotry_artifacts_read` | 发现与查看已生成的产物(异步交付 + 工作目录 markdown),行号文件视图,只读 |
 | **事实闸** | `gotry_fact_gate` | 行程产物交付前闸——见上文[工作原理](#工作原理)中的定义 |
 | **通用外部** | `gotry_web_search` · `gotry_video_subtitle` · `gotry_github_search` · `gotry_agent_reach` | 网页/字幕/GitHub/全渠道外部信息(经 Agent-Reach) |
+| **自检** | `gotry_doctor` | 可选依赖只读体检(扩展 / Agent-Reach .venv / hbcli / FlyAI key **+ 最近一次匿名试用达限时间** / dsh-calendar 挂载态 / sidebar),逐项状态 + 精确补装指引;安装只经用户在终端跑 `npx gotry doctor --fix` 完成。LLM key 归 dsh 宿主管,doctor 刻意不管。报告落 `gotry-state/doctor-report.md`(侧栏工作台可预览) |
+
+> **通道路由**:检索工具面保持平铺(无隐藏派发);persona 路由卡与检索失败结果内附的 `routing` 建议**由通道注册表单一生成**(官方 API > 用户会话 > 网页兜底,按会话健康面过滤)。某通道额度耗尽时,结果会明说并指名下一通道——盲重试是契约违例,不靠 prompt 碰运气。
 
 ## 一段对话
 
@@ -111,6 +114,21 @@ GoTry: 收到。先把约束记下来——
 
 > 标签导读:`[骨架:openflights]` 说的是"这条航线能飞"已被公开航线数据校验;`[实时API:*]` 说的是刚从实时接口拉回的当下数据;`[静态包:估算]` 提醒价格是淡季档估算——**订前核实**。标签由渲染层附加,模型无权染指。
 
+## 与主流 AI 同题横评(产品人格从哪来)
+
+GoTry 的产品人格不靠拍脑袋:同一段真实的跨国 workation 行程 prompt(埋了考点——不给年份、模糊指代「万xx」、用户已自行消解的歧义)逐字投喂各家主流 AI,回答逐字存档、对照地面真值打分,再反向校准行为契约。transcript、评分卡与契约反哺:[`docs/persona-bench/`](docs/evaluation/persona-bench/README.md)。
+
+| 维度 | 通用助手(Kimi,真实 13 轮) | OTA Agent(飞猪开放平台,单轮) | GoTry 契约 |
+|---|---|---|---|
+| 日历 grounding | ✗ 2025 年历;用户三次纠正、三次道歉式重排 | ✗ 派生星期落 2025 年历,与同页自答自相矛盾 | (2)(8)(9)+时间锚点卡 |
+| 约束访谈 | ✗ 零提问;两个要害约束第 6 轮才由用户说出 | △ 只问销售资质(预算/星级/靠海),必答题零命中 | (1)(10) |
+| 可行性与时间账 | ✗ 密度幻觉,被用户点破 | ✗ 香港办事+当天飞普吉无时间账;「约8小时」与自己的到达时刻矛盾 | (4)+门到门全成本 |
+| 事实可证性 | △ 目的地研究经得起对 | ✗ 停业航司(2020 年歇业)仍在售;价格全部无出处 | (3)(7)(13)(20) |
+| 结构完整性 | △ 第 13 轮才长出像样的对比表 | ✓✓ 单轮骨架最全——完整是基本盘 | 验证过的完整(事实闸) |
+| 人格一句话 | 博学但无状态的聊天者——用户被迫干四份工 | 版式完美的 OTA 导购——每段止于价格表 | 可信赖的行程工程师:访谈先行、求解器判决、不可行明说 |
+
+**单条最有价值的发现:两家互不相关的产品,要自己算的星期全落在 2025 年历上**——日历锚定必须是产品机制(锚点卡、一次断言、永不重算),不是模型运气。反面教材全文:[`docs/research/kimi-postmortem.md`](docs/research/kimi-postmortem.md)。
+
 ## 快速开始
 
 ### npm(推荐)
@@ -125,6 +143,7 @@ npx @danceiny/gotry web
 |---|---|---|
 | Web 对话(推荐) | `npx @danceiny/gotry web` | 持续多轮规划,看推理可视化 → `:3080` |
 | headless 一问一答 | `npx @danceiny/gotry "我想从深圳休整两天,预算 3000"` | 脚本 / CI / 定向调试 → stdout |
+| 依赖体检 | `npx @danceiny/gotry doctor`(`--fix` 补装) | 可选渠道不好使时:体检扩展 / Agent-Reach / hbcli / FlyAI key / sidebar,逐项给精确修复指引,报告落 `gotry-state/doctor-report.md`(侧栏工作台可预览) |
 
 前置:Node ≥ 22.15。LLM 凭证由 dsh 宿主 UI 配,OpenAI 兼容端点(MiniMax/中转/自建网关)走 dsh 的模型设置。首启 6–15 秒属正常冷启动;`:3080` 被占先腾端口;异常退出会留证据到 `gotry-state/incidents.jsonl`(不静默)。
 
@@ -165,23 +184,24 @@ node scripts/build-dist.mjs                       # 构建 JS runtime
 
 ## 状态与限制
 
-当前版本:**v0.0.1-rc.17**(npm dist-tag `rc`;`latest` 仍指 rc.16,沿用 dsh-管-LLM 安装口径)。评测处于 Phase 0 基座——确定性合同、校验器与节奏策略;无外部 benchmark 分数、无花费、无 uplift 声明。
+当前版本:**v0.0.1-rc.18**(npm `latest` 与 `rc` 均指本版;2026-09-03 registry 回拉实测:干净安装/bin/dist 入口全通)。评测处于 Phase 0 基座——确定性合同、校验器与节奏策略;无外部 benchmark 分数、无花费、无 uplift 声明。
 
 **今天可用**(全栈回归全绿;每项都有确定性测试):
 
 - **Z3 求解引擎** —— 可行性判决 + 门到门全成本;历史并发竞态已根治并进回归闸
-- **实时检索** —— 机票/火车/酒店(飞猪官方通道)、目的地/酒店目录、天气、航班观测、通航性校验;实时票价可覆写求解价(`GOTRY_REALTIME_PRICING=1`)
-- **账号会话检索** —— 你本人登录态查携程机票;观测轮次中所有可评分 hit 全过、ReadGuard 零写,非 hit 保持显式 `miss` 记录——不作超出此口径的实时可售声明
+- **实时检索** —— 机票/火车/酒店(飞猪官方通道)、目的地/酒店目录、天气、航班观测、通航性校验;实时票价可覆写求解价(`GOTRY_REALTIME_PRICING=1`);飞猪匿名试用额度达限归类 `needs-setup` 并带配 key 指引(不盲重试)
+- **依赖体检** —— `npx gotry doctor`(CLI)/ `gotry_doctor`(对话内工具):可选依赖(扩展 / Agent-Reach / hbcli / FlyAI key / sidebar)只读体检 + 精确补装指引;`--fix` 补装;LLM key 仍归 dsh 宿主管
+- **账号会话检索** —— 你本人登录态查携程机票/酒店 + 12306 火车(酒/火 2026-09-03 实装:酒店为被动嗅探登录态真实价,火车为 12306 公开余票查询面;接口面随首个真会话校准);观测轮次中所有可评分 hit 全过、ReadGuard 零写,非 hit 保持显式 `miss` 记录——不作超出此口径的实时可售声明
 - **扩展按需装** —— `[GoTry Session Bridge](https://chromewebstore.google.com/detail/gotry-session-bridge/oeajpiccmonococjcegddlooeeohlbgd)` 由 dsh 宿主 UI 在账号会话工具首次需要时以可点链接给出(Chrome 商店一键装 + 自动更新);gotry 这边不跑 setup wizard、不开 chrome://extensions、不动剪贴板
 - **记忆与触达** —— 动机画像 / 愿望池 / 同行人 / 旅行时间线;英文输出一键切换(`GOTRY_LOCALE=en`)
-- **有界的工具循环** —— 16 次派发后注入软收敛,18 次后结构化 `TOOL_BUDGET_EXHAUSTED` 拒绝;经打包消费者安装的 E2E 在 CI 里实测
+- **按任务路由的回合预算** —— 每轮先由确定性路由器(零 LLM)分类 quick / sync / deep-planning;时间才是唯一预算,且到点出口跟任务走:quick/sync 收敛作答,deep-planning 转后台落 `gotry_turn_handoff.v1` 工单(ETA 约 1 小时)而不是让会话流死掉;工单由 `scripts/turn-handoff-collect.ts` 在后台收集结算(幂等、带递归防护的子规划会话),回访时经只读工具 `gotry_turn_handoff_list` 查询状态与交付物;经打包消费者安装的 E2E 在 CI 里实测
 
 **已知限制**(诚实清单):
 
 - **M3 Exit 未关闭** —— 工程与分发面就绪,但真实种子用户证据(50–200 人 cohort)尚未积累;自动化测试证明的是合同与公式,不是 business pass
 - **酒店会话适配** —— 携程酒店/美团登录态面等实测回填;机票已通
 - **界面语言** —— 英文仅覆盖求解确定性输出层;dsh 宿主界面与对话面属宿主/校准件
-- **外部 benchmark 泛化** —— 迄今所有冻结外部运行均仅 diagnostic(无分数、无 uplift 声明);逐轮工程台账见 [`docs/benchmark-environment-bridge.md`](docs/benchmark-environment-bridge.md)
+- **外部 benchmark 泛化** —— 迄今所有冻结外部运行均仅 diagnostic（无分数、无 uplift 声明）。bridge 现已合并通用 `tools/call/errors` 恢复面、逐工具 typed descriptor、严格 result/domain/failure 分流与 latest-response terminal fence；逐轮工程台账见 [`docs/evaluation/benchmark-environment-bridge.md`](docs/evaluation/benchmark-environment-bridge.md)
 - **预订** —— 今天没有任何可下单路径;M5 只经 WriteGate 与 booking-saga 状态机启封
 
 <details>
@@ -235,17 +255,19 @@ npx tsx scripts/evaluation-cadence-tests.ts    # 确定性节奏策略/planner
 
 | 文档 | 内容 |
 |---|---|
+| [`docs/README.md`](docs/README.md) | 文档规范与总索引(目录税则 · 命名 · 生命周期) |
 | [`docs/architecture.md`](docs/architecture.md) | 系统 / ADR / 演进 / 债务清单(中文,权威) |
 | [`docs/gotry-master-outline.md`](docs/gotry-master-outline.md) | 总纲:工作分解 · 复用矩阵 |
 | [`docs/gotry-product-design.md`](docs/gotry-product-design.md) | 产品设计:主循环 · 透明机制 · 全成本模型 |
 | [`docs/roadmap.md`](docs/roadmap.md) | M0–M6 时间线与当前位置 |
 | [`docs/user-guide.md`](docs/user-guide.md) | 终端用户使用指南 |
 | [`docs/data-sources.md`](docs/data-sources.md) | 数据源与证据链政策 |
-| [`docs/extension-privacy.md`](docs/extension-privacy.md) | Session Bridge 扩展隐私 |
-| [`docs/benchmark-environment-bridge.md`](docs/benchmark-environment-bridge.md) | 外部 benchmark 桥——工程台账 |
-| [`docs/evaluation-foundation.md`](docs/evaluation-foundation.md) | 评测 Phase 0 基座 |
-| [`docs/booking-saga-fsm.md`](docs/booking-saga-fsm.md) | 预订 saga 状态机(M5 缝词汇) |
-| [`docs/kimi-postmortem.md`](docs/kimi-postmortem.md) | 一次真实 AI 旅行规划失败复盘(反面教材) |
+| [`docs/ops/extension-privacy.md`](docs/ops/extension-privacy.md) | Session Bridge 扩展隐私 |
+| [`docs/evaluation/benchmark-environment-bridge.md`](docs/evaluation/benchmark-environment-bridge.md) | 外部 benchmark 桥——工程台账 |
+| [`docs/evaluation/evaluation-foundation.md`](docs/evaluation/evaluation-foundation.md) | 评测 Phase 0 基座 |
+| [`docs/design/booking-saga-fsm.md`](docs/design/booking-saga-fsm.md) | 预订 saga 状态机(M5 缝词汇) |
+| [`docs/research/kimi-postmortem.md`](docs/research/kimi-postmortem.md) | 一次真实 AI 旅行规划失败复盘(反面教材) |
+| [`docs/evaluation/persona-bench/`](docs/evaluation/persona-bench/) | Agent 产品人格横评——同一真实行程 prompt 的各家回答存档、评分卡与人格提炼 |
 | [`docs/release-notes.md`](docs/release-notes.md) | 逐版本发布决策(「为什么」) |
 | [`CHANGELOG.md`](CHANGELOG.md) | 机器衍生的变更日志(Keep a Changelog + Conventional Commits) |
 | [`docs/tokens.md`](docs/tokens.md) | npm 2FA / 发布机制 |
@@ -258,7 +280,7 @@ npx tsx scripts/evaluation-cadence-tests.ts    # 确定性节奏策略/planner
 
 **Built with**: DeepSeek Harness 0.1.2-alpha.3 (root-pinned) · Cordis · Z3 (WASM) · loopx (pipx) · hotelbyte-cli · Agent-Reach v1.5.0 · OpenFlights · TypeScript
 
-**版本基线:`v0.0.1-rc.17`(npm dist-tag `rc`;`latest` 沿用 rc.16)。** 当前 checkout 的权威验证闸以 `scripts/run-all-tests.sh` 实际枚举为准;发布流程见 `scripts/publish-npm.sh`。
+**版本基线:`v0.0.1-rc.18`(npm `latest`)。** 当前 checkout 的权威验证闸以 `scripts/run-all-tests.sh` 实际枚举为准;发布流程见 `scripts/publish-npm.sh`。
 
 ---