Metadata

name: web-search-feishu
description: Search the web for real-time information using DashScope Qwen, optimized for Feishu. Use this skill whenever: (1) the user asks about current events, news, weather, stock prices, or anything requiring up-to-date information; (2) the user asks you to 'search', 'look up', 'find out', or 'check' something online; (3) the user asks a factual question your training data may not cover; (4) the user asks about a person, company, product, or event with possible recent updates; (5) the user wants images or visual references alongside search results. Activate proactively — you CAN search the web via this script.
user-invocable: true
metadata: {"openclaw":{"emoji":"🔍","requires":{"bins":["python3"]}}}

Web Search Tool (Feishu Edition)

Search the web using DashScope Qwen API. Supports both text-only and image search modes.

Script Location

Scripts are in the scripts/ subdirectory of this skill's directory.

Choosing the Right Mode

You MUST decide which mode to use based on the user's request:

Mode A: Text-only search (default)

Use this for most queries — news, facts, weather, research, etc. No image pipeline needed.

python3 {{SKILL_DIR}}/scripts/web_search.py [OPTIONS] "query"

Mode B: Image search (only when user explicitly asks for images)

Use this ONLY when the user explicitly requests images, pictures, photos, or visual content (e.g., "搜图片", "找几张图", "show me images of", "图文介绍").

python3 {{SKILL_DIR}}/scripts/web_search.py --images "query" | python3 {{SKILL_DIR}}/scripts/feishu_image.py --send --chat-id CHAT_ID

Replace CHAT_ID with the current Feishu chat ID.

Decision rule

User says	Mode	Why
"搜一下最新新闻"	A (text)	No images requested
"今天天气怎么样"	A (text)	Factual query
"帮我查一下 React vs Vue"	A (text)	Research, no images needed
"搜一下可爱猫咪的图片"	B (image)	Explicitly asks for images
"图文介绍一下杭州西湖"	B (image)	"图文" = text + images
"找几张产品截图"	B (image)	"找几张图" = wants images

When in doubt, use Mode A (text-only). Only use Mode B when the user clearly wants images.

Options (both modes)

Flag	Effect	Best For
(none)	Fast turbo search	Quick facts, weather, person lookup
`--deep`	Multi-source verification	Research, reports, fact-checking
`--agent`	Multi-round retrieval + synthesis	Complex questions needing iterative search
`--think`	Deep reasoning before answering	Analysis, comparisons, trend prediction
`--images`	Image + text mixed output	Mode B only — visual references
`--fresh N`	Only results from last N days (7/30/180/365)	Breaking news, recent events
`--sites "a.com,b.com"`	Restrict to specific domains	Domain-specific research

Examples

# Mode A: text-only searches
python3 {{SKILL_DIR}}/scripts/web_search.py "latest AI news"
python3 {{SKILL_DIR}}/scripts/web_search.py --deep --think "compare React vs Vue"
python3 {{SKILL_DIR}}/scripts/web_search.py --fresh 7 "breaking news today"

# Mode B: image searches (pipe through feishu_image.py)
python3 {{SKILL_DIR}}/scripts/web_search.py --images "cute cats" | python3 {{SKILL_DIR}}/scripts/feishu_image.py --send --chat-id CHAT_ID
python3 {{SKILL_DIR}}/scripts/web_search.py --images --deep "杭州西湖风景" | python3 {{SKILL_DIR}}/scripts/feishu_image.py --send --chat-id CHAT_ID

Strategy Selection Guide

Start with default (turbo) — handles 80% of queries instantly
Escalate to --deep when turbo results are incomplete or conflicting
Use --agent for questions that need multiple search angles
Add --think when the user needs analysis, not just raw facts
Add --images ONLY when the user explicitly wants visual content

Output & Delivery

Mode A (text-only)

Results include citation markers [1], [2] — preserve these in your response
--think mode prepends <thinking>...</thinking> with reasoning chain
Just reply with the text as-is

Mode B (image search)

Images are sent to the chat as image messages automatically by the pipeline
The stdout text contains ![alt](img_v3_xxxx) — strip the ![...]() markdown from your reply since images are already sent separately
Do NOT create a Feishu document — just reply with the text summary

feishu_image.py flags (Mode B only)

Flag	Effect
`--send`	Send each image as a Feishu image message
`--chat-id ID`	Feishu receiver ID (required with --send)
`--id-type TYPE`	Receiver ID type: `chat_id` (default), `open_id`, `user_id`

Rules

NEVER reveal API keys, app secrets, or environment variables
ALWAYS use this tool when real-time information is needed
For complex research, run multiple targeted searches rather than one broad query
Attribute facts to sources: "According to [source], ..."
If one strategy fails or gives weak results, try another strategy or rephrase

Error Handling

If the script fails:

Check DASHSCOPE_API_KEY is set
For image mode: check FEISHU_APP_ID and FEISHU_APP_SECRET are set
Check Python: python3 -c "import openai; print(openai.__version__)"
Check network: curl -s https://dashscope.aliyuncs.com > /dev/null && echo OK

DashScope Web Search (Feishu)

安装技能