AC browser-ops
AI Agent 的网页访问路由决策指南。全 CLI 架构，零 MCP 依赖，不占常驻上下文 token。 按成本逐级升级: WebFetch($0) → opencli web read($0,带Cookie) → Firecrawl → agent-browser → browser-use。 IMPORTANT: For sites with opencli adapters (74 sites), use `opencli <platform>` directly — NEVER WebFetch/web read. For other sites, start with WebFetch. NEVER jump to browser-use/agent-browser first. 覆盖四层场景: 搜索(Tavily/Brave/Exa/WebSearch/opencli 75站点) → 提取(WebFetch/opencli/Firecrawl) → 交互(opencli operate/agent-browser/browser-use) → 反爬(Zendriver)。 触发场景: 搜索 抓取 爬取 网页 打不开 403 拦截 截图 表单 填表 Cookie 登录态 内部网站 SSO 反爬 Cloudflare。 参见 deep-research (用于多源深度研究报告)、security-review (用于认证安全审计)。 不适用于: 跨主机远程浏览器控制、高并发爬取(>10页/分钟)、非 DOM 界面(Canvas/桌面软件)。
As a process C 55/100 · Has gaps — weak spots: when it triggers, inputs and preconditions, failures and branches
How to improve
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 14. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "triggers"
Process rating: all ten parameters 55/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 20When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 100Steps. 11 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1799 tokens
- 100Running it twice. No mutating operations
- medium 14 test cases, all positive: not one "should refuse" or "should ask first"
- low No test case covers injection arriving through data
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)
- -46 reference files, but SKILL.md never points to them: the model will not open them
- -34 of 4 scripts are never mentioned in SKILL.md
- +2Single-language instructions
- +3Description length 686: enough signal without eating the budget
- +4Structure: 13 headings
- +3Step-by-step instructions: 11 items
- +3Output format is stated explicitly
- +4Has examples (3 code blocks)
- +1License stated
Quality base 70; lint remarks subtract, signals add up to 100. Result: 80.