This is a test.
`; const $ = cheerio.load(html); console.log($("p.info").text()); ``` -------------------------------- ### List available scripts in SKILL.md Source: https://agentskills.io/skill-creation/using-scripts Reference bundled files using relative paths from the skill directory root in your SKILL.md file to make them available to the agent. ```markdown ```markdown SKILL.md # My Skill ## Scripts - run-linter: scripts/linter.sh - format-code: scripts/formatter.sh ``` ``` -------------------------------- ### Run npm packages with npx Source: https://agentskills.io/skill-creation/using-scripts Use npx to run npm packages, downloading them on demand. It is bundled with Node.js. Pin versions for reproducibility. ```bash npx eslint@9 --fix . ``` ```bash npx create-vite@6 my-app ``` -------------------------------- ### SKILL.md File Structure Source: https://agentskills.io/what-are-skills The SKILL.md file contains YAML frontmatter for metadata and Markdown for instructions. Required frontmatter includes 'name' and 'description'. ```mdx --- name: pdf-processing description: Extract PDF text, fill forms, merge files. Use when handling PDFs. --- # PDF Processing ## When to use this skill Use this skill when the user needs to work with PDF files... ## How to extract text 1. Use pdfplumber for text extraction... ## How to fill forms ... ``` -------------------------------- ### Define procedural vs specific instructions Source: https://agentskills.io/skill-creation/best-practices Contrast between hardcoded specific answers and reusable procedural methods for agent tasks. ```markdown Join the `orders` table to `customers` on `customer_id`, filter where `region = 'EMEA'`, and sum the `amount` column. 1. Read the schema from `references/schema.yaml` to find relevant tables 2. Join tables using the `_id` foreign key convention 3. Apply any filters from the user's request as WHERE clauses 4. Aggregate numeric columns as needed and format as a markdown table ``` -------------------------------- ### Run Ruby Script Source: https://agentskills.io/skill-creation/using-scripts Execute a Ruby script using the `ruby` command. This assumes Bundler is available and configured correctly for inline gems. ```bash ruby scripts/extract.rb ``` -------------------------------- ### Use Structured Output Formats Source: https://agentskills.io/skill-creation/using-scripts Prefer structured formats like JSON, CSV, or TSV for output. This allows for programmatic consumption by agents and standard command-line tools. Separate data from diagnostics by sending structured data to stdout and messages to stderr. ```bash # Whitespace-aligned — hard to parse programmatically NAME STATUS CREATED my-service running 2025-01-15 # Delimited — unambiguous field boundaries {"name": "my-service", "status": "running", "created": "2025-01-15"} ``` -------------------------------- ### Validating skills with `skills-ref` Source: https://agentskills.io/specification Use the `skills-ref validate` command to check if your `SKILL.md` frontmatter is valid and adheres to naming conventions. ```bash skills-ref validate ./my-skill ``` -------------------------------- ### Run scripts with deno run Source: https://agentskills.io/skill-creation/using-scripts Use deno run to execute scripts directly from URLs or specifiers. Requires permission flags for filesystem/network access and uses '--' to separate Deno flags from tool flags. ```bash deno run npm:create-vite@6 my-app ``` ```bash deno run --allow-read npm:eslint@9 -- --fix . ``` -------------------------------- ### Define Client Data Structure Source: https://agentskills.io/clients An array of objects containing metadata for each agent product, including branding assets and documentation links. ```javascript export const clients = [{ name: "Junie", description: "Junie is an LLM-agnostic coding agent built for real-world development. It is built on top of the IntelliJ Platform, so it understands your project the same way your editor does.", url: "https://junie.jetbrains.com/", lightSrc: "/images/logos/junie/junie-logo-on-white.svg", darkSrc: "/images/logos/junie/junie-logo-on-dark.svg", instructionsUrl: "https://junie.jetbrains.com/docs/agent-skills.html" }, { name: "Gemini CLI", description: "Gemini CLI is an open-source AI agent that brings the power of Gemini directly into your terminal.", url: "https://geminicli.com", lightSrc: "/images/logos/gemini-cli/gemini-cli-logo_light.svg", darkSrc: "/images/logos/gemini-cli/gemini-cli-logo_dark.svg", instructionsUrl: "https://geminicli.com/docs/cli/skills/", sourceCodeUrl: "https://github.com/google-gemini/gemini-cli" }, { name: "Autohand Code CLI", description: "Autohand Code CLI is an autonomous LLM-powered coding agent that lives in your terminal. It uses the ReAct (Reason + Act) pattern to understand your codebase, plan changes, and execute them with your approval.", url: "https://autohand.ai/", lightSrc: "/images/logos/autohand/autohand-light.svg", darkSrc: "/images/logos/autohand/autohand-dark.svg", scale: 0.8, instructionsUrl: "https://autohand.ai/docs/working-with-autohand-code/agent-skills.html", sourceCodeUrl: "https://github.com/autohandai/code-cli" }, { name: "OpenCode" ``` -------------------------------- ### Define output format templates Source: https://agentskills.io/skill-creation/best-practices Provide a structured template to ensure consistent output formatting from the agent. ```markdown ## Report structure Use this template, adapting sections as needed for the specific analysis: ```markdown # [Analysis Title] ## Executive summary [One-paragraph overview of key findings] ## Key findings - Finding 1 with supporting data - Finding 2 with supporting data ## Recommendations 1. Specific actionable recommendation 2. Specific actionable recommendation ``` ``` -------------------------------- ### POST /_mintlify/feedback/agent-skills/agent-feedback Source: https://agentskills.io/home Submits feedback regarding incorrect, outdated, or confusing documentation. ```APIDOC ## POST /_mintlify/feedback/agent-skills/agent-feedback ### Description Submit feedback for specific and actionable issues found in the documentation. ### Method POST ### Endpoint https://agentskills.io/_mintlify/feedback/agent-skills/agent-feedback ### Parameters #### Request Body - **path** (string) - Required - The current page path where the issue was found. - **feedback** (string) - Required - Description of the issue. ### Request Example { "path": "/current-page-path", "feedback": "Description of the issue" } ``` -------------------------------- ### Skill Test Cases Configuration Source: https://agentskills.io/skill-creation/evaluating-skills Define your skill's test cases in `evals/evals.json`. Each test case includes a prompt, expected output, and optional input files. This structure allows for systematic testing and iteration. ```json { "skill_name": "csv-analyzer", "evals": [ { "id": 1, "prompt": "I have a CSV of monthly sales data in data/sales_2025.csv. Can you find the top 3 months by revenue and make a bar chart?", "expected_output": "A bar chart image showing the top 3 months by revenue, with labeled axes and values.", "files": ["evals/files/sales_2025.csv"] }, { "id": 2, "prompt": "there's a csv in my downloads called customers.csv, some rows have missing emails — can you clean it up and tell me how many were missing?", "expected_output": "A cleaned CSV with missing emails handled, plus a count of how many were missing.", "files": ["evals/files/customers.csv"] } ] } ``` -------------------------------- ### Define code review process guidelines Source: https://agentskills.io/skill-creation/best-practices A markdown-based checklist for performing code reviews, focusing on security and concurrency. ```markdown ## Code review process 1. Check all database queries for SQL injection (use parameterized queries) 2. Verify authentication checks on every endpoint 3. Look for race conditions in concurrent code paths 4. Confirm error messages don't leak internal details ``` -------------------------------- ### Record run timing and token data Source: https://agentskills.io/skill-creation/evaluating-skills JSON structure for capturing performance metrics like total tokens and duration in milliseconds upon task completion. ```json { "total_tokens": 84852, "duration_ms": 23332 } ``` -------------------------------- ### Implement ClientShowcase React Component Source: https://agentskills.io/clients A functional component that manages client display state and renders a responsive grid of cards. Requires an array of client objects as props. ```javascript export const ClientShowcase = ({clients}) => { const shuffle = arr => { const copy = arr.slice(); for (let i = copy.length - 1; i > 0; i--) { const j = Math.floor(Math.random() * (i + 1)); [copy[i], copy[j]] = [copy[j], copy[i]]; } return copy; }; const sort = clients => clients.toSorted((a, b) => a.name.localeCompare(b.name)); const [state, dispatch] = React.useReducer((state, action) => { switch (action) { case "shuffle": return { mode: "shuffle", clients: shuffle(clients) }; case "alpha": return { mode: "alpha", clients: sort(clients) }; default: return state; } }, { mode: "shuffle", clients: shuffle(clients) }); const Logo = ({client}) =>{client.description}
{(client.instructionsUrl || client.sourceCodeUrl) &&This is a test.
' print(BeautifulSoup(html, "html.parser").select_one("p.info").get_text()) ``` -------------------------------- ### Run Validation Script Source: https://agentskills.io/skill-creation/using-scripts Execute the validation script with a specified input file. Ensure the script path is relative to the skill directory root. ```bash bash scripts/validate.sh "$INPUT_FILE" ``` -------------------------------- ### Wrap Skill Content with Identifying Tags Source: https://agentskills.io/client-implementation/adding-skills-support Use identifying tags to wrap skill content when using a dedicated activation tool. This helps the model distinguish skill instructions and allows the harness to identify skill content during context compaction. ```xmlThis is a test.
`; const $ = cheerio.load(html); console.log($("p.info").text()); ``` -------------------------------- ### Define a roll-dice skill in SKILL.md Source: https://agentskills.io/skill-creation/quickstart Create this file in .agents/skills/roll-dice/SKILL.md to define the skill's name, description, and execution instructions. ```markdown --- name: roll-dice description: Roll dice using a random number generator. Use when asked to roll a die (d6, d20, etc.), roll dice, or generate a random dice roll. --- To roll a die, use the following command that generates a random number from 1 to the given number of sides: ```bash echo $((RANDOM %