CD webclaw3
浏览器自动化 / 网页数据采集 / 爬虫抓取——用你自己的 Chrome 和已登录状态干活，不用重新登录、不用交账号密码、不怕反爬。所有需要操作浏览器的活都走这个 skill。 场景一（日常上网干活）：搜东西、看网页、查数据、抓列表、导表格、下载内容，或者在网站里动手操作——点按钮、填表单、发内容、和商家聊天；包括必须登录才能看的页面，如小红书、抖音、微博、推特(X)、知乎、B站、公众号、淘宝、拼多多、Shopify 后台等。 场景二（沉淀成 skill）：任务跑通了，用户说"帮我提炼""以后每天自动跑"，把这次探索固化成可反复运行、零 token 的本地脚本 skill，可配定时任务。 场景三（修复）：网站改版导致之前生成的 skill 跑失败了，用户来找修，本地直接修好。 (EN) Browser automation and web scraping that drives your own logged-in Chrome — explore once, distill it into a free, deterministic skill you can rerun daily.
浏览器自动化 / 网页数据采集 / 爬虫抓取——用你自己的 Chrome 和已登录状态干活，不用重新登录、不用交账号密码、不怕反爬。所有需要操作浏览器的活都走这个 skill。…
As a process D 44/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions
What is at stake
The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.
The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".
If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The skill contains commands that delete files, rewrite disks or execute code fetched from the network. The agent may run them without asking if it believes the instructions require it.
Replace destructive commands with safe equivalents that ask for confirmation, scope them to one folder, and stop piping curl into a shell: pin a version and a checksum.
How to improve
- Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 5
-
high Exfiltration
intent-browser-credential-storeskill-card.md:17Accesses a browser credential / cookie storeDevelopers and agent users use webclaw3 to perform browser automation, authenticated web data extraction, page interaction, and recurring workflow generation from their own Chrome session. It is espec
Medium and low: 4
-
medium Dangerous commands
cmd-pipe-to-shellreferences/setup.md:20Downloads and executes remote code from an unrecognised host (pipe to shell) (documentation table row)| `qoderwork-cn` | **QoderWork 国内**（PC 桌面） | `qoderclicn` | `curl -fsSL https://qoder.cn/install \| bash` | https://docs.qoder.cn/cli/qoder-cli-cn-get-started-quickly |
table -
medium Dangerous commands
cmd-pipe-to-shellscripts/webclaw3.mjs:41Downloads and executes remote code from an unrecognised host (pipe to shell) (string literal in code, not executed)'qoderwork-cn': { type: 'qoder-code', binary: 'qoderclicn', label: 'QoderWork 国内', doc: 'curl -fsSL https://qoder.cn/install | bash' },code literal -
medium Dangerous commands
cmd-install-from-urlscripts/webclaw3.mjs:304Installs a package from an untrusted URL / archive (code comment)// 底层执行 npm i -g <skillDir>/dist/wc3-…ne-*.tgz（离线安装，无外部依赖）
comment -
medium Dangerous commands
cmd-install-from-urlscripts/webclaw3.mjs:345Installs a package from an untrusted URL / archive (code comment)// 手动兜底（供排障，不打给普通用户）：npm i -g <skillDir>/dist/wc3-…ne-*.tgz
comment
Files scanned: 22. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - note
frontmatter-keyunknown frontmatter key "github"
Process rating: all ten parameters 44/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 3 mutating operations with no state check
- 40Consistency. Frontmatter name (webclaw3) differs from the folder (webclaw3-browser-automation)
- 50Failures and branches. 0 branches, has a failure section
- 60Tools and files. Uses tools (bash, web, node) that frontmatter does not declare
- 100Steps. 53 steps
- 100Execution cost. Instruction body is 3916 tokens
- low 11 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Output format is not stated: the model decides each time
- -2localhost URLs: will not work for another user
- -34 of 5 scripts are never mentioned in SKILL.md
- +2Single-language instructions
- +3Description length 502: enough signal without eating the budget
- +4Structure: 24 headings
- +3Step-by-step instructions: 53 items
- +4Has examples (8 code blocks)
- +4Reference files are cited in the instructions (4 of 12)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 71.