Skillful AI Public Docs v1
SkillfulAI_Docs_v1.0
SkillfulAI_Docs_v1.0
  • Overview of Skillful AI
  • Introduction
    • AI Models
      • Types of AI models
    • Decentralization of AI
    • LLM Architectures
    • Current Challenges
    • Our Solution
  • Ecosystem Overview
  • Product Suite
    • Model Hub
      • Collaborative Mode
    • AI Builder
      • 1. Creating a Character
      • 2. Language Model
      • 3. Strategy
      • 4. Skills
        • List of Skills
      • 5. Memory
      • 6. Integrations
      • 7. Agent Info & Tasks
    • Agent Hub
    • AI Marketplace
    • Image Hub
    • Wallet
  • Tokenomics
  • Pricing
  • Links
  • Glossary
  • About Us
  • FAQs
  • Partnerships
Powered by GitBook
On this page
  1. Product Suite
  2. AI Builder
  3. 4. Skills

List of Skills

Previous4. SkillsNext5. Memory

Last updated 15 days ago

Coinmarketcap:

  • convert_price - Perform currency conversions across fiat () and cryptocurrencies. It can convert one symbol at a time. If the request is to convert 7 coins, then 7 individual tool calls occur.

    • How to Request Data:

      • You can either request conversion using Coinmarketcap ID or Symbols:

        • Using ID

          • Provide CoinMarketCap ID of the source currency (Example → "1,2" for Bitcoin and Liteccoin) and target conversion currency (Example → 1 for Bitcoin).

            • Note: IDs are different from CoinMarketCap's rankings. Rankings refer to the list of coins you see by default when you open CoinMarketCap’s website (i.e., sorted by market cap). On the other hand, CoinMarketCap assigns a unique ID to each coin when it is first listed on the platform. This ID, which remains constant over time, is part of their internal system for reliably tracking and converting data.

          • Examples:

            • Convert crypto ID 1 to 2781 (2781 represents USD as fiat)

            • Convert crypto ID 1 to 2 (2 represents the cryptocurrency symbol LTC)

          • Using Symbol:

            • Provide the ticker symbol of the source currency (Example → "BTC, ETH") and target fiat or cryptocurrency symbols to convert the amount into.

            • Examples:

              • Convert 1 BTC to USD (to fiat)

              • Convert 1 BTC to ETH (crypto currency symbol)

  • get_cryptocurrency_info - Retrieve comprehensive details (static metadata) about specific cryptocurrencies like name & launch date, description, logo, use & ecosystem, website URL, supply, price (with last 24h performance), 24-hour trading volume, contract address, social links, source code, explorer and links to official docs.

    • How to Request Data

      • You need to provide at least one of the following:

        • ID (CoinMarketCap ID): Example → Get info about crypto id 1

        • Slug (Crypto name): Example → Get info about Bitcoin

        • Symbol (Ticker symbol): Example → Get info about BTC

        • Contract Address: Example → Get info about 0xc40af1e4fecfa05ce6bab79dcd8b373d2e436c4e

Here's what you can expect as a output from this skill:

  • get_cryptocurrency_listings_latest - Returns a paginated list of all active cryptocurrencies with latest market data.

    • There isn't any max limit on the number of tickers/ID a user request in one call. It depends on Agent performance. On an average, users can expect 10 records per request.

    • By default, it will use coin's market cap as a filter. However, you may sort against any one of the following (per request):

      • market_cap: CoinMarketCap's market cap rank

        • Example: List the top 10 cryptocurrencies by market cap.

      • market_cap_strict: A strict market cap sort (latest trade price x circulating supply)

        • Example: Show the top 10 cryptocurrencies strictly sorted by market cap calculation.

      • name: The cryptocurrency name.

        • Example: Show cryptocurrencies sorted by name in ascending order.

      • symbol: The cryptocurrency symbol.

        • Example: Show cryptocurrencies sorted by their ticker symbol.

      • date_added: Date cryptocurrency was added to the system.

        • Example: List the most recently added cryptocurrencies

      • price: latest average trade price across markets.

        • Example: Find cryptocurrencies priced between $1 and $100.

      • circulating_supply: approximate number of coins currently in circulation.

        • Example: What is the current circulation of BTC?

      • total_supply: approximate total amount of coins in existence right now (minus any coins that have been verifiably burned).

        • Example: What is the total supply of coins between ID 20 and 30?

      • max_supply: approximation of the maximum amount of coins that will ever exist in the lifetime of the currency.

        • Example: Find cryptocurrencies sorted by their max supply.

      • num_market_pairs: number of market pairs across all exchanges trading each currency.

        • Example: Show cryptocurrencies with the highest number of trading pairs across exchanges

      • market_cap_by_total_supply_strict: market cap by total supply.

        • Example: List cryptocurrencies with the highest market cap based on total supply.

      • volume_24h: rolling 24 hour adjusted trading volume.

        • Example: Show cryptocurrencies that increased by at least 2% yesterday between marketcap of 100M and 1B

      • volume_7d: rolling 24 hour adjusted trading volume.

        • Example: Show cryptocurrencies with the highest trading volume over the past 7 days.

      • volume_30d: rolling 24 hour adjusted trading volume.

        • Example: Show cryptocurrencies with the highest trading volume over the past 30 days.

      • percent_change_1h: 1 hour trading price percentage change for each currency.

        • List cryptocurrencies that have increased by at least 2% in the past hour.

      • percent_change_24h: 24 hour trading price percentage change for each currency.

        • Example: List cryptocurrencies that gained at least 10% in the last 24 hours.

      • percent_change_7d: 7 day trading price percentage change for each currency.

        • Example: List cryptocurrencies that gained at least 15% in the last 7 days.

You can mix and match any of the above skills.

  • Example: Show the cryptocurrencies strictly sorted by market cap (latest price × circulating supply), with prices between $1 and $100, market cap between $100M and $1B, and at least a 2% gain in the last 30 days.

NOTE: Use this endpoint if you need a sorted and paginated list of all cryptocurrencies. If you want to query for market data on a few specific cryptocurrencies use get_cryptocurrency_quotes_latest which is optimized for that purpose. The response data between these endpoints is otherwise the same.

  • get_cryptocurrency_listings_new - Returns a sorted and paginated list of most recently added cryptocurrencies.

    • Note that the listings are limited to 10 records per request.

    • You can also convert to fiat currency or cryptocurrency.

    • Example: Get new listing of cryptocurrency and convert the prices to ETH

  • get_cryptocurrency_ohlcv_historical - Fetch historical OHLCV (Open, High, Low, Close, Volume) data for cryptocurrencies.

    • Time Intervals are set based on duration

      • If the requested duration is 1 or more than 1 year, data is retrieved for only any 2 months (User needs to specify these 2 months).

      • If the requested duration is 1 or more than 1 month, data is retrieved weekly.

      • If the requested duration is 1 or more than 1 week, data is retrieved daily.

      • If the requested duration is 1 or more than 1 day, data is retrieved hourly.

        • Technical Notes

          • Date Format: Use an ISO date format like "YYYY-MM-DD" (e.g., "2018-09-19") for daily data.

          • If you want fewer data points, you can set interval (e.g., "2d" means data every 2 days instead of every day).

          • If you don’t specify time_start, it will get data counting backward from time_end.

            • Example Prompt: "Get hourly OHLCV data for Ethereum for the last 24 hours." Since the user did not provide a time_start, the system would automatically calculate it by going back 24 hours from the current time. The time_end would be the current time at the moment of the request. The count would be 24 to retrieve 24 data points, each representing one hour.

            • To get OHLCV data for a specific date, set time_start to one day before. For example, to get data for November 30, 2024, use November 29, 2024.

            • There are two ways to set time intervals: i) Time Period and ii) Interval

              • Time Period: It uses fixed UTC-based periods, such as "hourly", "daily", "weekly", "monthly" and "yearly". Daily data is available 5-10 minutes after midnight UTC. Hourly data is available 5 minutes after each hour.

                • Example: Get daily OHLCV data for Bitcoin form 2nd March to 5th March 2025.

              • Intervals: For example, "1h", "2h", and "3h" return data every 1, 2, or 3 hours, while "1d", "2d", and "7d" provide data every 1, 2, or 7 days.

                • Example: Get OHLCV data every 2 days for the last week etc.

            • You can adjust data frequency. For daily data, use "1d". For every other day, use "2d".

              • Example user prompt: "Get OHLCV data for Bitcoin for every 2d in March 2025."

            • You can also combine both time intervals:

              • Example: Get OHLCV data every 6 hours for Bitcoin for the last 2 days, Get OHLCV data every 2d for Ethereum from March 1, 2025, to April 30, 2025

  • get_cryptocurrency_ohlcv_latest - Retrieve the latest OHLCV data for cryptocurrencies for the current UTC day. Since the current UTC day is still active, these values are updated frequently. Update frequency is every 10 minutes.

    • How to Request Data

      • You need to provide at least one of the following:

        • Id: The CoinMarketCap ID of the cryptocurrency.

        • Slug: The URL-friendly identifier of the cryptocurrency.

        • Symbol: The ticker symbol of the cryptocurrency.

        • Convert: The currency code for price conversion.

  • get_cryptocurrency_quotes_latest - Gives the latest market quote for 1 or more cryptocurrencies. This include price, 24h trading Volume, market cap, circulating supply, total supply, recent price changes etc.

    • How to Request Data

      • You need to provide at least one of the following:

        • Id: The CoinMarketCap ID of the cryptocurrency.

        • Slug: The URL-friendly identifier of the cryptocurrency.

        • Symbol: The ticker symbol of the cryptocurrency.

        • Convert: The currency code for price conversion.

NOTE: Use this endpoint to request the latest quote for specific cryptocurrencies. If you need to request all cryptocurrencies use get_cryptocurrency_listings_latest which is optimized for that purpose. The response data between these endpoints is otherwise the same.

  • get_exchange_info - Obtain detailed information about cryptocurrency exchanges such as launch date, logo, official website URL, social links, and market fee documentation URL.

    • How to Request Data

      • You need to provide at least one of the following:

        • ID: The CoinMarketCap ID of the exchange.

        • Slug: The URL-friendly identifier of the exchange (i.e., name of exchanges like "coinbase", "binance" etc.)

  • get_fear_and_greed_latest - Access the latest "Fear and Greed Index" to gauge market sentiment.

    • Since the current UTC day is still active, these values are updated frequently. Update frequency is every 15 minutes.

  • get_fear_and_greed_historical - Access paginated list of all CMC Crypto Fear and Greed values at 12am UTC time. The data is available for last 60 days. Updates every 15 seconds.

  • get_global_metrics_quotes_historical - Returns an interval of historical global cryptocurrency market metrics based on time and interval parameters. This includes:

    • Bitcoin Dominance,

    • Ethereum Dominance,

    • Total Active Cryptocurrencies,

    • Total Active Exchanges,

    • Total Active Market Pairs,

    • Total Crypto 24 Hour Volume,

    • Altcoin Market Cap,

    • Altcoin 24 Hour Volume

      • Technical Notes

        • The API supports different time intervals depending on the requested timeframe:

          • Hourly (hourly): Used for durations over 1 day and up to 7 days.

            • Example: Example: Get Bitcoin dominance for the last 4 days.

          • Daily (daily): Used for durations over 7 days and up to 30 days.

            • Example: Get Bitcoin dominance for the last 20 days.

          • Weekly (weekly): Used for durations over 30 days and up to 365 days.

            • Example: Get Total Crypto 24 Hour Volume for last 2 months.

          • Monthly (monthly): Used for durations over 1 year (upto 24 months but you must request it in two-month intervals per request).

            • Example: Find altcoin 24 Hour Volume for entire 2023 year.

        • Interval Options

        There are 2 types of time interval formats that may be used for "interval".

        • The first are calendar year and time constants in UTC time:

          • "hourly" - Get the first quote available at the beginning of each calendar hour.

          • "daily" - Get the first quote available at the beginning of each calendar day.

          • "weekly" - Get the first quote available at the beginning of each calendar week.

          • "monthly" - Get the first quote available at the beginning of each calendar month.

          • "yearly" - Get the first quote available at the beginning of each calendar year

        • The second are relative time intervals.

          • "m": Get the first quote available every "m" minutes (60 second intervals). Supported minutes are: "5m", "10m", "15m", "30m", "45m".

          • "h": Get the first quote available every "h" hours (3600 second intervals). Supported hour intervals are: "1h", "2h", "3h", "4h", "6h", "12h".

          • "d": Get the first quote available every "d" days (86400 second intervals). Supported day intervals are: "1d", "2d", "3d", "7d", "14d", "15d", "30d", "60d", "90d", "365d".

        • If you provide both a start time and an end time, the API returns quotes at your chosen interval (e.g., hourly, daily). If you don’t provide start time, the API works backward from end time.

          • Example: Get Bitcoin dominance for the last 2 days. (Since start time is not provided, the API will fetch data backward from the current time)

        • If end time is not supplied, it defaults to the current time.

          • Example: Get Bitcoin dominance every 6 hours for the 3 days. Since the end time is not supplied, it defaults to the current time.

  • get_global_metrics_quotes_latest - Access current global market metrics for cryptocurrencies. Updates every 5 minute. The information includes (you can convert below market values in multiple fiat and cryptocurrency conversions in the same request):

    • Bitcoin Dominance,

    • Ethereum Dominance,

    • Total Active Cryptocurrencies,

    • Total Active Exchanges,

    • Total Crypto Market Cap

    • Total Crypto 24 Hour Volume,

    • Total Active Market Pairs,

    • Market Cap Change (24h % change)

    • Volume Change (24h % change)

    • DeFi Segment's marketcap and 24h Volume

    • Stablecoin segment's marketcap and 24h Volume

    • Derivatives segment 24h volume with a 24h zsnipercentage change

Image Generator:

  • skillful image generator - Generate Images from Prompt

    • Optional Settings: You can leave these out if you're not sure — the defaults will still generate good images. But if you want more control, here’s what each one does:

      • Inference steps – Controls the level of detail. Higher steps make the image more polished and clear. (Range: 25–100, default: 50)

      • Guidance scale – Controls how closely the image should match your prompt. Higher value means the image will follow your description more strictly. (Range: 1.0–30.0, default: 10.1)

      • Seed value – Use this if you want to generate the same image again later. Same prompt + same seed = same result. (Range: 1–1000, default: 1)

      • Comic style – Say you want a “comic-style” image if you want it to look like a drawing or illustration. If you don’t mention it, the image will be in a regular style by default.

Social Media:

  • telegram_send_message - Lets your AI agent send a message to someone on Telegram — either in a 1-on-1 DM, a group, or a public channel. You can also reply to a specific message if you have the message ID.

    • Users need to provide below things to use TG Skill:

      • Chat ID

      • Text of the message

      • Optional: A message ID to reply to a specific message

What are Telegram IDs?

There are 3 main types of IDs you might need when sending messages:

  • For sending a direct message (DM) to a person, you’ll need their user ID, which is just a number like 1234567890.

  • For sending a message to a private group, you’ll use a group ID. These usually start with -100, like -1009988776655.

  • For posting in a public channel, you can either use the channel's username like @newsupdates or its numeric ID, which also starts with -100.

What You Want to Do

What You Should Use

Example

Send a message to someone in DM

Use their numeric chat ID only (usernames won’t work in DMs)

“Send ‘Hey, how are you?’ to chat ID 1234567890”

Send a message to a Telegram group Group is a two-way chat where multiple members can message and interact.

Use the group’s numeric chat ID (usually starts with -100)

“Tell the group ‘Reminder: stand-up call in 10 mins.’ Use chat ID -1009988776655”

Post in a public Telegram channel Channel is a one-way broadcast tool where only admins can post updates to subscribers.

Use either the channel username (e.g., @companychannel) or the numeric channel ID

“Send ‘Update deployed. Details in the doc.’ to @companychannel”

Reply to a message in a DM

Use both chat ID and message ID you’re replying to

“Reply in chat ID 1234567890 to message 456: ‘Thanks for the update!’”

Reply in a group chat

Same as above: you’ll need the group’s chat ID and the message ID

“Reply in group ID -1009988776655 to message 789: ‘Good suggestion, I’ll add that.’”

Advanced Settings

This setting controls who the Telegram bot is allowed to reply to. For private chats (DMs), you can choose to: let no one get replies, let everyone message the bot, or allow only specific people (whitelist) using their username or user ID. Similarly, for group chats, you can either let anyone in the group mention the bot and get a reply, or restrict it to only selected users from your whitelist. In short, it’s a privacy and control feature to limit bot access only to trusted users if needed.

You can find the advanced settings option under the Integration tab of Agent hub.

  • x_twitter_create_post: This lets you post a new tweet on your connected X (Twitter) account. It can publish a new tweet, reply to another tweet, or attach an image to a tweet.

    • You must include at least one of the following:

      • text: The tweet content (words, emojis, hashtags, etc.)

      • image: A public image URL to attach to the tweet

      • reply_to_id: The tweet ID you want to reply to (only used when replying)

    • Use Cases (How You Might Use It)

      • Post on X: ‘Just published a new blog — check it out

      • Tweet this image: https://example.com/promo.jpg

      • Tweet: ‘We’re live now 🚀’ with this image: https://site.com/banner.png

      • Reply to tweet ID 17983722: ‘Appreciate the mention!’

      • Reply to tweet 17983722 and say: ‘Thanks! Here’s the new version.’ with image: https://site.com/new.png

  • x_twitter_search_posts - This lets your agent search recent public tweets based on any custom query. You can filter tweets by hashtags, mentions, usernames, language, and more — and get real-time or historical results.

    • What You Need to Provide:

      • Query (required): It's the search string that tells the agent what tweets to find. You can use a combination of below filters to refine your query.

        • Basic Filters

          • #: Hashtag

            • Finds tweets containing the hashtag #OpenAI.

          • @: Mention.

            • Finds tweets mentioning @TwitterDev.

          • Tweets from a specific user.

            • Shows tweets from @elonmusk.

          • Tweets replying to a specific user.

            • Finds tweets replying to @TwitterDev.

          • Filter retweets.

            • Shows only retweets with #AI

          • Filter replies.

            • Shows tweets that are replies and mention @elonmusk

          • Filter quote Tweets (tweets that share another tweet with a comment).

            • Shows quote tweets that mention #crypto

          • Tweets in a specific language.

            • Shows only English tweets

          • Exclude a term or operator.

            • Finds tweets with #AI but excludes retweets

      • Logical Operators

        • Find tweets that match both conditions at the same time

          • Find tweets that include the hashtag #OpenAI and are written in English

        • Find tweets that match either one condition or the other

          • Find tweets that include either the hashtag #AI or #MachineLearning

        • Group parts of a query, especially when using OR

          • Find tweets that are either from @OpenAI or include the hashtag #OpenAI

      • Start time (optional): This sets how far back in time you want to search. It must be after 21st march 2006.

        • Get tweets with the hashtag Bitcoin, starting from April 1, 2025

      • End time (optional): This sets the latest date to search up to. Also in UTC format. If you don’t include it, the agent will search up to the current moment.

        • Get tweets with the #Solana, only up to April 5, 2025

      • Max results (optional): The number of tweets you want in the result. Must be between 10 and 100. If you don’t set it, it defaults to 10.

        • Get 50 tweets that include #AI

      • Next token (optional): If your search returns too many tweets, you can set max results to limit how many you get at once (like 50), and then use next token to ask for the next batch.

        • Give me the next batch of tweets about #Ethereum after the first 100

      • Sort order (optional): You can choose between "recency" or "relevancy". The default is "recency", which means the newest tweets show up first. Relevancy refers to tweets most relevant to your query, not necessarily the newest.

        • Show the newest tweets with #AI

        • Show the tweets most closely related to #AI

Stock Market & Company Analysis:

  • find_similar_companies - Identify and compare companies with similar profiles or metrics.

    • Filtering Criteria:

      • It finds similar companies related to a specified ticker, based on news coverage and returns data. This helps users identify peers, competitors, or thematically related companies.

    • Output format: Company ticker

    • Example: Find similar companies to $AAPL

  • get_current_aggressive_small_cap_stocks - Screen aggressive small-cap stocks with high growth potential. You will always get the first 25 stocks that fits into the filtering criteria.

    • Filtering Criteria:

      • 1-year EPS (earnings per share) change greater than 25%

      • Marketcap (intraday) less than $2B

    • Output format: Stock symbol, market capitalization, and P/E (TTM) (Price-to-Earnings Trailing Twelve Months) ratio

      • P/E (TTM) ratio [Price-to-Earnings Trailing Twelve Months] calculates a company's stock price relative to its earnings per share (EPS) over the past 12 months.

  • get_current_gainer_stocks - Discover stocks with the largest gains in the market. You will always get the first 25 stocks that fits into the filtering criteria.

    • Filtering Criteria:

      • A daily price increase of more than 3%

      • A stock price of at least $5

      • A minimum marketcap of $2B

      • A trading volume of over 15,000 shares

    • Output format: Stock symbol, marketcap and P/E (TTM) ratio

  • get_current_loser_stocks - Identify stocks with the most significant losses. You will always get the first 25 stocks that fits into the filtering criteria.

    • Filtering Criteria:

      • The stock’s price change must be less than –2.5%

      • The stock’s price must be at least $5

      • A minimum marketcap of $2B

      • The trading volume must exceed 20,000 shares

    • Output format: Stock symbol, marketcap and P/E (TTM) ratio

  • get_current_most_traded_stocks - Find the most actively traded stocks. You will always get the first 25 stocks that fits into the filtering criteria.

    • Filtering Criteria:

      • Total number of shares traded during the current trading day—that is, from when the market opens (9:30 AM ET) to when it closes (4:00 PM ET).

    • Output format: Stock name and symbol, P/E ratio, marketcap and 52-week price % change.

  • get_current_stock_price_info - Fetch real-time stock price information and details.

    • Output format: Stock name and symbol, current Price (USD), high Price of the day, low price of the day, open price of the day and price percentage change and time.

  • get_current_technology_growth_stocks - Screen technology sector stocks with growth potential. You will always get the first 25 stocks that fits into the filtering criteria.

    • Filtering Criteria:

      • Stocks in technology sector only

      • A Year-over-Year EPS growth of more than25%

      • A quarterly revenue growth of more than 25%

    • Output format: Stock name and symbol, marketcap and P/E Ratio (TTM).

  • get_current_undervalued_growth_stocks - Identify undervalued growth stocks for potential investments. You will always get the first 25 stocks that fits into the filtering criteria.

    • Filtering Criteria:

      • A Price / Earnings ratio (trailing) between 0 and 20

      • A Price/Earnings to Growth (PEG) ratio of less than 1

      • A minimum of 25 % change in Earning Per Share (Basic)

    • Output format: Stock name and symbol, marketcap and P/E Ratio (TTM).

  • get_current_undervalued_large_cap_stocks - Screen large-cap stocks with strong value indicators.

    • Filtering Criteria:

      • A marketcap between $10B - $100B

      • P/E ratio (trailing) below 20

      • A Price/Earnings to Growth (PEG) ratio of less than 1

    • Output format: Stock name and symbol, marketcap and P/E Ratio.

  • get_earnings_history - Gives annual and quarterly earnings (EPS) for the company of interest. Quarterly data also includes analyst estimates and surprise metrics.

    • Output format: Quarterly earnings history including the Fiscal Date Ending, Reported Date, Estimated EPS, Actual EPS, EPS Surprise, and Surprise Percentage.

  • get_latest_earning_estimate - Fetch the latest earnings estimates for stocks.

    • Output format: Stock name with symbol and EPS estimate

  • get_stocks_with_upcoming_earnings - Discover stocks scheduled to release earnings within a chosen time period (e.g., next 2 days, 1.5 weeks, 3 months, etc.)

    • Output format: Stock name and symbol, Report date

News & Trends Skills:

  • get_google_trending_searches - Analyze trending searches on Google.

    • The trending searches are centered on topics relevant to the last 24 hours in United States.

    • You can get top 3 trending searches per request.

  • get_google_trends_for_query - Retrieve trend data for specific queries.

    • You will get top 5 "Rising Query" and Top 5 "Top query" related to your specific query.

      • Each of these 5 query will include name, % search increase volume in last 24 hours (value), and a link of google trend search for more information.

        • The "Rising Queries" are queries related to that query term that have had significant increases over a recent period.

          • Example: For the main query "tornado warning", top queries might include terms like "Tornado Warning California Scotts Valley", "Tornado Warning Finleyville PA", "San Francisco Tornado Warning" etc.

        • The "Top Queries" are other common search terms associated with that specific query.

          • A trend may consist of multiple queries that are variants of the same search or considered to be related. Top Queries details these queries.

          • Example: For the main query "tornado warning", Top Queries might include terms like "Tornado Watch", "Weather Tornado Warning", "Warning vs Watch", "tornado warning columbus ohio" etc.

  • get_latest_news_for_stock - Access the latest news articles about specific stocks.

    • You will get one line summary for each news and a link for more information.

  • get_trending_finance_news - Stay updated with trending financial news stories.

    • You will get news heading along with 1 line summary and a link to read more.

Web & Data Collection Skills:

  • basic_web_scraper.py - Enable basic web scraping to collect and process data from web pages/URLs. Users can ask for either the plain text (cleaned-up readable content) or the full HTML source of the page.

  • citation - Enable generation of properly formatted citations based on the topic you give. You can choose the citation style, such as APA, MLA, or Chicago. User needs to provide a topic and citation style.

    • Examples:

      • Give me a citation on climate change in APA style

      • I need an MLA citation for artificial intelligence

      • Generate a Chicago-style citation for the history of the internet

      • Can you create a citation for quantum computing using APA?

Solana:

    • Note that only one token can be queried per request.

    • Mention the token ticker to receive information.

    • Example: Show token details for $BONK.

  • solscan Get Account Token List - Retrieves a list of tokens owned by an account.

    • You can request either tokens or NFTs for a particular account on the Solana chain.

    • Note that only 10 tokens will be shown in the output.

    • You can sort the tokens based on their USD value in ascending or descending order.

    • Example: Give tokens held by [address] in descending order based on USD amount of each token.

  • solscan Get Token Holders - fetch a list of wallet addresses that hold a specific token on Solana.

    • You must provide the token's contract address as input.

    • You can filter holders based on the minimum or maximum token holding amount.

    • Note that the output may include contract addresses, hot/cold wallets, and exchange addresses.

    • Example: List holders of token [token address] with a minimum of 10,000 tokens.

  • solscan Get Token List - Retrieves a list of tokens from the Solscan Pro API.

    • The list can be sorted in ascending or descending order, filtered by one of the following:

      • Number of token holders, or

      • Market cap, or

      • Token creation time

    • Example: Show top 20 tokens sorted by market cap in descending order.

  • solscan Get Token Metadata - Retrieves the metadata of a token. Please provide the token address (and not the token ticker) to receive the metadata.

    • The metadata includes the token's name, symbol, description, website, social links (e.g., Twitter), total supply, current price, 24-hour volume, market cap, market cap rank, 24-hour price change, holder count, mint authority address (and first mint transaction), freeze authority (if any), and creator address (and its creation transaction).

    • Example: Fetch metadata for [token address]

  • solscan Get Token Price - Retrieves the historical or current price of a token. Please provide the token address to get the price.

    • Example: Get historical daily prices of [token address] from January 1, 2025.

  • solscan Get Top Token List - Retrieves a list of top tokens.

    • You can filter top token by name, symbol, market cap, creation time, price, 24 hour price change, and number of holders.

      • Example: Show the top tokens filtered by their 24 hour positive price change in descending order.

  • solscan Get Trending Token List - Retrieves a list of trending tokens. This include name of token, symbol and its address.

    • Example: What are the current top trending tokens on Solana?

Pump.fun

  • pump_fun_creation_token - Allows AI agent to create and deploy a new token on the Solana blockchain.

    • What You Need to Provide:

      • Required Fields:

        • Name – The full name of your token. Example: FunnyCoin

        • Symbol – The short symbol for your token. Example: FNY

        • Image – The file path or link to your token image. Example: "path/to/logo.png"

      • Optional Settings (Quick Guide)

        • Description – A short description of what your token is about. Example: “A meme token built for fun on Solana.”

        • Twitter link – X page for your token or project. Example: https://twitter.com/FunnyCoinSOL

        • Telegram link – A Telegram group or channel for your token community. Example: https://t.me/FunnyCoinGroup

        • Website – The official website for your token. Example: https://funnycoin.io

        • Show name publicly – If you want the token name to be shown on supported platforms, say something like: “Yes, show the token name publicly” (If not mentioned, it will stay hidden by default.)

        • Amount (in SOL) – The amount of SOL used for the token creation. Default is 0.1 Example: “Use 0.25 SOL for the transaction”

        • Slippage – The maximum allowed price slippage during the creation process (in %). Default is 10 Example: “Set slippage to 5 percent”

        • Priority fee – An optional fee (in SOL) to speed up the transaction. Default is 0.0005 Example: “Add 0.001 as priority fee for faster processing”

      • Example:

  • pump_fun_trade_token - Allows AI agent to buy or sell tokens on the Solana blockchain using PumpFun.

    • What You Need to Provide:

      • Required Fields:

        • Action – What you want to do: either buy or sell

        • Mint address – The unique identifier for the token on Solana (usually a long string) Example: "7GxQ...abc" (paste the actual token mint address)

        • Amount – How much you want to trade.

          • You can use a fixed number like "100.5" or a percentage like "100%"

        • Denominated in SOL – Let the agent know if the amount is in SOL or in the token itself Example:

          • I want to spend 0.5 SOL to buy

          • I want to sell 50 tokens

      • Optional Settings

        • Slippage – How much price movement you're okay with before canceling the trade. Lower slippage is safer but might fail in fast markets. (Default is 10%)

        • Priority fee – A small fee in SOL to speed up the transaction. (Default is 0.0005 SOL) Example: Add a priority fee of 0.001 SOL

        • Pool – Choose the liquidity pool to trade from. Options: "pump" (default) or "raydium" Example: Use the Raydium pool instead of Pump

      • Examples:

        • Buy 0.8 SOL worth of tokens from this mint: 7GxQXei9yAnUXN5CDeeYdX2PBvVgwsPz2UpPFPVdLabc, with 3% slippage and a priority fee of 0.002 SOL.

        • Buy 150 tokens of E2oRvFAzNz7hZJhS74KtkM9K6J6FbFzRSvRAVWMPq123, use Raydium, allow 2% slippage, and add 0.0008 SOL as priority fee

Weather Services:

  • weather - Access real-time weather information for various locations

Media Processing:

  • It uses the Gemini model to handle a variety of media types, including images, videos, and audio files. It supports direct URLs to media files (e.g., .jpg, .png, .mp4, .mp3) and links to platforms like YouTube.

    • Image Processing: When given a link to an image or screenshot, the model can:

      • Extract text from within the image.

      • Identify visual elements such as objects, people (excluding private identification), animals, and environments.

      • Summarize what the image shows and describe the overall scene.

    • Video Processing: For video links, including those from YouTube, the model can:

      • Analyze both the visual frames and audio track.

      • Provide a summary of the video’s content.

      • Identify key actions, scenes, objects, or speakers (if audio is clear).

    • Audio Processing: When provided with an audio file link, the model can:

      • Transcribe spoken words into text.

      • In some cases, separate speakers based on voice differences.

      • Recognize important audio cues, such as music, sound effects, or tone changes.

solana Token Search - Retrieves tokens matching a given symbol from the This includes the token's name, logo, description, daily trading volume, verification status on Solscan, creation date, and mint date.

Make a token called MemeKing, symbol MKING, with description ‘The king of memes’, and link it to our Twitter page: , use 0.3 SOL, set slippage to 5%, add a priority fee of 0.001, and show the name.

93 major currencies
Jupiter.
https://twitter.com/memeking”