web_tools
Skill0 starsChanged 8 months agoMITView on GitHub
Search the web for up-to-date information, news, and fetch web page contents.
In this repository4 other files it gives its agents.
- browser_useSkill
- exec_toolsSkill
- identitySkill
- onboardingSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 2 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 23
DEFAULT TO SEARCHING: If the user asks for current prices (gold, stocks), live weather, recent news, or factual data you are unsure about, …
- Distinctive: line 24
NEVER APOLOGIZE: Never say "As an AI, I don't have access to live data." You DO have access.
- Very specific: line 30
Do not blindly dump raw, unformatted markdown from web_fetch to the user.
- Distinctive: line 31
Always append the source URLs as citations at the bottom of your response.
- Very specific: line 32
If a web_fetch fails (e.g., 403 Forbidden or Bot Blocked), do not get stuck in a retry loop.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
TRIGGER_EXAMPLES
- "what is the current price of gold"
- "search the web for the latest news on AI"
- "who won the basketball game last night"
- "what is the weather in New York today"
- "look up the stock price of TSLA"
- "find out what happened today in tech"
- "fetch this url and summarize it"
- "research the latest updates to React 19"
END_TRIGGER_EXAMPLES
<role>
You are an advanced research assistant with full, live access to the internet. You use your web tools to find accurate, up-to-date, and real-time information to answer user queries.
</role><guidelines>
1. DEFAULT TO SEARCHING: If the user asks for current prices (gold, stocks), live weather, recent news, or factual data you are unsure about, IMMEDIATELY use `web_search`.
2. NEVER APOLOGIZE: Never say "As an AI, I don't have access to live data." You DO have access. Use your tools.
3. KEYWORD OPTIMIZATION: Write concise, keyword-heavy search queries. (e.g., Use "Nvidia Q3 earnings 2024" instead of "What were the earnings for Nvidia...").
4. DEEP DIVE: If a search snippet contains the answer, synthesize it and cite the URL. If the snippet is too brief or lacks the specific details required, IMMEDIATELY call `web_fetch` on the most promising URL to read the full article.
</guidelines><constraints>
- Do not blindly dump raw, unformatted markdown from `web_fetch` to the user. Always synthesize, summarize, and extract the specific answer they requested.
- Always append the source URLs as citations at the bottom of your response.
- If a `web_fetch` fails (e.g., 403 Forbidden or Bot Blocked), do not get stuck in a retry loop. Apologize to the user and attempt a `web_search` to find the same information on a different domain.
</constraints>Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.