AgentSkillsCN

Gen2WebSearch

通过DuckDuckGo搜索，以强大的内容提取总结链接结果页面，并按域名/日期使用CommonCode FolderNavigator记录日志。

SKILL.md
--- frontmatter
name: Gen2WebSearch
description: Search DuckDuckGo, summarize linked result pages with robust content extraction, and save logs by domain/date using CommonCode FolderNavigator.
metadata: {"openclaw":{"requires":{"bins":["python","python3"]},"os":["win32","linux","darwin"]}}

Gen2WebSearch

Combines search from LoggedWebSearchPlus with page extraction quality from Gen2WebText, and domain-aware folder resolution via CommonCode FolderNavigator.

Execution

python gen2_web_search.py <domain> --query <search_query> --max <number> --timeout <seconds> --page-timeout <seconds> --words <number> --sleep-ms <milliseconds> --date-context <none|month_year|year> --date <YYYY-MM-DD>

Arguments

  • •domain required alphabetic string only A-Z or a-z
  • •--query required search query
  • •--max optional default 8 range 1 to 25
  • •--timeout optional default 20 range 5 to 60
  • •--page-timeout optional default 15 range 5 to 60
  • •--words optional default 200 range 80 to 400
  • •--sleep-ms optional default 0
  • •--date-context optional default none. Appends date hints to the query using natural tokens:
    • •none uses the query as-is
    • •month_year appends <mon> <yyyy> (example feb 2026)
    • •year appends <yyyy>
  • •--date optional reference date in YYYY-MM-DD. Used only to derive date-context tokens; if omitted, today is used.

Output

On success returns JSON with domain, query, effective_query, date_context, query_reference_date, count, results, log_path, status. Each result includes title, url, snippet, rank, page_summary, page_words, page_status.

On error returns JSON with domain, query, error, results, count, status.

Logging

Writes to: {OPENCLAW_ROOT}/data/datastore3/01-Mine/<domain>/YYYY/MM/DD/{QueryName}.txt

File name is derived from the query text. Path resolution is handled by FolderNavigator.from_fixed_point() and get_today_path(area="mine", domain=..., create=True).