Docs for Lite 2026.10.19, core 0.73.0 and ThinkWatch 3.8.0

src/content/docs-core/en/overview.md

@@ -15,9 +15,9 @@ Clients such as Claude Code and Codex send their requests to the gateway, and Co
 - **Rule-based routing.** Rules match on the model, the gateway key, the input size, the presence of tools or images and other properties of a request, and send it to an upstream or a group, rewrite its parameters or refuse it. When the client and the upstream use different API formats, the request is converted between Anthropic Messages, OpenAI Chat Completions, OpenAI Responses and Gemini.
 - **Failover before the first byte.** Until the first byte reaches the client, a failing upstream is replaced by the next one without the client noticing. After that point, the failure is reported. A circuit breaker keeps requests away from an upstream that keeps failing.
 - **Cost accounting.** Token usage and cache hits are priced from a public price table, which the control plane refreshes daily, or from a price sheet in the configuration. Each request records its cost and where the price came from. Estimated amounts are marked as such, and usage that cannot be priced is labelled *unknown* rather than given an invented figure.
-- **Outbound redaction.** Credentials and personal information in a request are replaced with placeholders before the request leaves, and restored when the model echoes them back. Field names are not searched, and custom rules also skip the ids, types, roles and names a protocol needs as they are.
-- **Tool-call inspection.** Tool calls returned by an upstream are checked against a rule set, and a dangerous call can be cut off mid-stream.
-- **Content filter.** The user messages and tool results a client sends are checked by keyword, regular expression or code point, and each rule refuses the request, deletes what it matched or only records it; the built-in rules also catch hidden characters that can carry instructions. Each of the three guards is set to `off`, `observe` or `enforce`, and all start in `observe`.
+- **Outbound redaction** (Secrets & privacy in ThinkWatch Lite). Credentials and personal information in a request are replaced with placeholders before the request leaves, and restored exactly as written when the model echoes them back, tool-call arguments included; a rule can instead only record what it finds. Field names are not searched, and custom rules also skip the ids, types, roles and names a protocol needs as they are.
+- **Tool-call inspection** (Dangerous commands in ThinkWatch Lite). Tool calls returned by an upstream are checked against a rule set, and a dangerous call can be cut off mid-stream or only recorded.
+- **Content filter.** The user messages and tool results a client sends are checked by keyword, regular expression or code point, and a rule can also check the text of the answer as it streams. Each rule refuses the request, deletes what it matched, replaces it with a text of its own or only records it; the built-in rules also catch hidden characters that can carry instructions. Each of the three guards is set to `off`, `observe` or `enforce`, and all start in `observe`; under `enforce` every rule does what its action says.
 - **An encrypted control plane.** The desktop app and `twcore` commands reach core over a local socket (a loopback port on Windows) and, when it is enabled, a remote control port. Every control connection starts with a Noise handshake keyed by `listen.control.key`; there are no certificates.
 
 ## Further reading

src/content/docs-core/zh-CN/overview.md

@@ -15,9 +15,9 @@ Claude Code、Codex 等客户端把请求发往网关后，Core 提供以下功
 - **按规则路由。** 规则按模型、网关密钥、输入规模、是否携带工具或图片等请求属性匹配，将请求发往某个上游或策略组、改写其参数，或拒绝请求。客户端与上游的接口格式不同时，请求在 Anthropic Messages、OpenAI Chat Completions、OpenAI Responses 与 Gemini 之间转换。
 - **首字节前的故障转移。** 首字节到达客户端之前，出错的上游由下一个上游替换，客户端无从察觉；此后发生的故障如实报告。熔断器使持续出错的上游暂不接收请求。
 - **费用核算。** token 用量与缓存命中按公开价目表计价（控制面每天刷新一次），或按配置中的价目表计价；每个请求都记录费用及价格来源。估算的金额另行标注，无法计价的用量标记为「未知」，不会填入虚构的数值。
-- **出站脱敏。** 请求发出之前，其中的凭据和个人信息替换为占位符；模型回显时再恢复原值。字段名不在查找之列，自定义规则还会跳过协议要求原样保留的 id、类型、角色和名称。
-- **工具调用审查。** 上游返回的工具调用按规则集审查，高危调用可在流式传输中途截断。
-- **内容过滤。** 客户端发送的用户消息和工具结果按关键词、正则表达式或码位检查，每条规则可拒绝请求、删除命中的内容或仅记录；内置规则也能查出可夹带指令的隐藏字符。三项防护各可设为 `off`、`observe` 或 `enforce`，出厂均为 `observe`。
+- **出站脱敏**（ThinkWatch Lite 中称为「密钥与隐私」）。请求发出之前，其中的凭据和个人信息替换为占位符；模型回显时按原值的写法如实恢复，工具调用参数中的值也是如此。规则也可以只记录、不替换。字段名不在查找之列，自定义规则还会跳过协议要求原样保留的 id、类型、角色和名称。
+- **工具调用审查**（ThinkWatch Lite 中称为「危险命令」）。上游返回的工具调用按规则集审查，高危调用可在流式传输中途截断，或只作记录。
+- **内容过滤。** 客户端发送的用户消息和工具结果按关键词、正则表达式或码位检查，规则也可以在回答流式输出时检查回答正文。每条规则可拒绝请求、删除命中的内容、替换为指定的文字或仅记录；内置规则也能查出可夹带指令的隐藏字符。三项防护各可设为 `off`、`observe` 或 `enforce`，出厂均为 `observe`；`enforce` 下每条规则按其动作处理。
 - **加密的控制面。** 桌面应用与 `twcore` 命令通过本地 socket（Windows 上为回环端口）连接 core，开启后也可经远程控制端口连接。每条控制连接都先以 `listen.control.key` 完成 Noise 握手，不使用证书。
 
 ## 后续阅读

src/content/docs-lite/en/compare-cc-switch.md

@@ -24,7 +24,7 @@ A dash means the feature is not described in that product's documentation.
 | API format conversion | Proxy mode: Claude Code to OpenAI Chat Completions or Responses; Codex to Chat Completions or Anthropic Messages | Among Anthropic Messages, OpenAI Chat Completions, OpenAI Responses and Gemini |
 | Usage and cost | Requests, tokens, cache hit rate and estimated cost, from proxy logs or the clients' session logs; custom prices, optional sync from models.dev; quota and balance display | Tokens, cost and requests by period, model and upstream; LiteLLM prices refreshed daily, or custom price sheets; estimates marked, unpriced requests counted separately; the balances and quotas that upstreams report |
 | Request details | Provider, model, tokens, cost, timing and status; parameters, a response summary and errors | Matched rule, each attempt with its status, request and response bodies, cost; full-text search; replay against another upstream |
-| Outbound redaction and tool-call inspection | — | API keys, private keys, ID and bank card numbers replaced before a request leaves; tool calls that download and run code or send out credentials cut off. Both start by only recording |
+| Secrets & privacy and Dangerous commands | — | API keys, private keys, ID and bank card numbers replaced before a request leaves; tool calls that download and run code or send out credentials cut off. Both start in Monitor, which records and notifies |
 | MCP and skills | One MCP server list synced to the selected clients; skills installed from GitHub or ZIP files | MCP servers of 13 clients side by side, copied or removed for 4; skills and hooks listed; all scanned for hidden characters, prompt injection, dangerous commands and overly broad permissions |
 | Prompts | Prompt presets written to `CLAUDE.md`, `AGENTS.md` or `GEMINI.md` | — |
 | Sessions | Reads the clients' own session files; search, resume in a terminal, delete | Requests that passed through the gateway grouped into sessions and replayed turn by turn |