desirecore/market
2
0
Fork 0
mirror of https://git.openapi.site/https://github.com/desirecore/market.git synced 2026-10-23 15:13:53 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix(web-access): support system browser startup and offline readiness (#223)

Browse Source 
## 变更 / Changes

补充系统已安装浏览器的显式路径审批启动说明，区分原生 launched 回执与 CDP
自动化就绪。高级交互使用客户端从随包固定资产准备的解释器，不再尝试系统 Python 或联网
pip/npm。新参数仅在客户端实际声明支持时使用；浏览器仍需部署方另行安装并遵守许可。

Document approved executable-path startup for system-installed browsers,
distinguish native process creation from CDP readiness, and use the
client's offline bundled interpreter for advanced interaction. New
parameters require an advertising client. Browser installation and
licensing remain operator prerequisites.

## 验证 / Validation

中英正文、发现描述与版本 3.7.0 同步；结构/sidecar 校验无错误，翻译检查通过。公共敏感信息扫描通过。平台对应修复另以主仓 PR
提交。
This commit is contained in:
谢源祥
2026-10-22 19:18:50 +08:00
committed by GitHub
parent a514b55b68
commit 4e9cc94ca2
3 changed files with 46 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files

37
skills/web-access/SKILL.md
View File

@@ -15,7 +15,7 @@ description: >-
新闻、网址、URL、找一下、搜一下、查一下、小红书、B站、微博、飞书、Twitter、
推特、X、知乎、公众号、已登录、登录状态。
license: MIT
version: 3.6.0
version: 3.7.0
compatibility: Requires a DesireCore build with the bundled external-browser Playwright runtime; Chrome/Edge/Chromium remains separately installed. Builds advertising BrowserExternalLaunch can start it through tool approval in an isolated profile; older builds require manual CDP startup.
type: procedural
risk_level: low
@@ -45,7 +45,7 @@ provides:
- LocalBookmarks
metadata:
author: desirecore
updated_at: '2026-10-14'
updated_at: '2026-10-22'
i18n:
default_locale: en-US
source_locale: zh-CN
@@ -55,16 +55,16 @@ metadata:
zh-CN:
name: 联网访问
short_desc: 联网搜索、网页抓取、内置受管浏览器登录态访问与取文、研究调研工作流
description: 联网访问工具包——搜索公开页面、Jina 优化抓取、内置受管浏览器完成登录态访问与取文，以及用户点名时接管他自己的 Chrome/Edge/Chromium。 外部浏览器需单独安装并满足其使用条款；提供启动工具的客户端可经审批启动，旧客户端需手动开启调试。
description: 联网访问工具包——搜索公开页面、Jina 优化抓取、内置受管浏览器完成登录态访问与取文，以及用户点名时接管他自己的 Chrome/Edge/Chromium。 外部浏览器需单独安装并满足其使用条款；提供启动工具的客户端可经审批启动，旧客户端需手动开启调试。 支持新客户端按显式路径审批启动系统浏览器；原生启动与 CDP 自动化就绪分别核验，依赖从随包资产离线准备。
body: ./SKILL.zh-CN.md
source_hash: sha256:a81f7a90824259f0
source_hash: sha256:5a3b56b621b6edc3
translated_by: human
en-US:
name: Web Access
short_desc: Web search, page fetching, logged-in access via the governed built-in browser, research workflows
description: A web-access toolkit — search public pages, fetch heavy pages via Jina Reader, reach and read logged-in sites through the governed built-in browser, and drive the user's named Chrome/Edge/Chromium over CDP on request. The external browser is separately installed and subject to its usage terms; clients with the launch tool support approved startup, while older clients require manual debugging startup.
description: A web-access toolkit — search public pages, fetch heavy pages via Jina Reader, reach and read logged-in sites through the governed built-in browser, and drive the user's named Chrome/Edge/Chromium over CDP on request. The external browser is separately installed and subject to its usage terms; clients with the launch tool support approved startup, while older clients require manual debugging startup. New clients support approved native launch by explicit executable path; native startup and CDP automation readiness are verified separately, with automation dependencies prepared from bundled offline assets.
body: ./SKILL.md
source_hash: sha256:a81f7a90824259f0
source_hash: sha256:5a3b56b621b6edc3
translated_by: human
market:
icon: >-
@@ -160,9 +160,17 @@ If any fetch fails, explicitly tell the user which URL failed and which fallback
## Prerequisites: external browser + CDP (L3-external only)
**Only needed when taking the L3-external route.** The built-in browser has no prerequisites.
### System-installed browsers and offline readiness
### Always probe before attaching
Only clients whose declared Launch schema includes system and executablePath accept `BrowserExternalLaunch(browser="system", executablePath=<authorized absolute browser executable>, url=...)` for any locally installed browser. Use OS registration facts from `BrowserExternalProbe(requestedBrowser="any")` or an authorized executable path; never infer absence from the Chrome/Edge fixed-directory list alone. Approval must identify the actual executable. Operators must separately install the browser and comply with its licence and usage terms.
`system` returns `launched` with `automation="native"`: this proves process creation only, not navigation or automation readiness. Choose an available approved desktop-control method and verify the real window; do not pass this receipt to BrowserAct/BrowserScript or claim task completion. For compatible custom Chromium installations, clients accepting `executablePath` can launch the selected executable with its matching CDP family and isolated profile. Inspect the receipt and fresh Probe before interaction. Never silently replace a named product.
For advanced CDP interaction, use Probe's returned `automationRuntime.pythonExecutable`. The client prepares its interpreter and wheels from fixed bundled assets independently of Hatch/Volta and system Python. If unavailable, report the structured deployment error. Do not try system Python, online pip/npm installation or an invented internal session. Missing packaged assets require installation repair; opening a login page is not evidence that automation is ready.
**The remainder of this prerequisite section applies only to the CDP route.** Native system launch follows the section above and does not require a ready CDP endpoint or a Python runtime. The built-in browser has no external prerequisites.
### CDP only: always probe before attaching
Map the user's words to an exact probe request:
@@ -176,14 +184,14 @@ Handle the structured result exactly:
| status | Required response |
|---|---|
| `ready` | For a simple open/navigation, call `BrowserExternalOpen` with the returned exact browser id and port. For advanced click/read/extract interaction, continue with isolated Playwright `connect_over_cdp`. Name the detected external browser honestly. For `any`, the already-ready endpoint is the prepared choice even if other products are installed |
| `browser_not_installed` | Say the requested browser was not detected. If `alternatives` is non-empty, ask whether the user wants one of them; **never switch automatically** |
| `browser_not_installed` | Say the requested browser was not detected; this does not prove absence. Use authorized read-only OS registration/shortcut inspection for nonstandard installations. If its executable is found, use the supported explicit-path launch mode. Ask before substituting another product unless the user already authorized that choice |
| `debug_port_closed` | On a client advertising `BrowserExternalLaunch`, call it with the exact chosen browser, port and URL through the existing approval flow; probe again after success. Do not ask the user to type a command. On older clients without this tool, offer the manual display-only fallback |
| `browser_choice_required` | No endpoint is ready and multiple external browsers are installed. List id/name only and ask which one the user wants; probe that exact choice next |
| `browser_choice_required` | Multiple installations need a selection. If the user delegated the choice, select a listed installation; otherwise ask. CDP selections use an exact probe; system selections use the returned executable path for approved native launch |
| `browser_mismatch` | Say which browser is actually on the port and which one was requested; ask the user to correct the port or explicitly approve the other browser |
| `invalid_cdp_endpoint` | Explain that something is listening on the port but it is not a valid Chrome DevTools endpoint; do not attach |
| `host_unavailable` | Explain that this Agent Service cannot inspect the user's desktop host; do not assume a browser is installed or silently use the built-in browser |
`BrowserExternalProbe` stays read-only. `BrowserExternalLaunch` is the approved effectful entry:
For CDP mode, `BrowserExternalProbe` stays read-only and `BrowserExternalLaunch` is the approved effectful entry:
use the user's exact browser id, the selected loopback port and requested HTTP(S) URL. It launches
only the detected native executable with a client-owned isolated profile, waits for verified CDP,
and returns `opened` with `started`, process id and profile when a new process was started.
@@ -204,9 +212,9 @@ from “browser installed but debugging disabled”, and a random HTTP service m
⚠️ When attached over CDP, **never call `browser.close()`** — that would close the user's own external browser.
Only close the page you opened. Full recipes in [references/cdp-browser.md](references/cdp-browser.md).
### Simple external open/navigation: use the dedicated tool
### CDP only: simple external open/navigation
When the user's requested action is only to open or navigate to an HTTP(S) URL in their named external browser:
For the CDP route, when the requested action is only to open or navigate to an HTTP(S) URL:
1. Call `BrowserExternalProbe` and require `ready`.
2. In the same runtime session, call `BrowserExternalOpen` with `browser: detectedBrowser.id`, the exact probe `port`, and the URL.
@@ -246,6 +254,9 @@ the missing prerequisite. Do not improvise a cross-shell command and do not fall
```
User intent
│
├─ Other installed browser / explicit executable path
│ └─→ Inspect installation facts if needed; advertised system Launch → native desktop control and real-window verification. No CDP-ready grant or Python prerequisite
│
├─ **Any request that names "my own / my machine's / external Chrome/Edge/Chromium"**
│ └─→ L3-external first, whether the verb is search/read/open/click: