.codex-plugin/plugin.json

{
  "name": "aya-web-extract",
  "version": "1.0.0",
  "description": "Turn permitted public HTML listings into source-linked datasets you can filter and export.",
  "author": {
    "name": "Aya Studio LLC",
    "email": "contact@yourfriendaya.com",
    "url": "https://yourfriendaya.com"
  },
  "homepage": "https://extract.yourfriendaya.com",
  "keywords": [
    "web extraction",
    "datasets",
    "research",
    "CSV",
    "jobs",
    "products"
  ],
  "skills": "./skills/",
  "mcpServers": "./.mcp.json",
  "interface": {
    "displayName": "Aya Web Extract",
    "shortDescription": "Public pages to datasets",
    "longDescription": "Extract repeated records from permitted public HTML product listings, job boards, and directories. Choose fields in plain language, inspect every record with its source link, filter the saved dataset, and export CSV or JSON. Jobs are limited to 10 pages and 250 records, and datasets expire after 7 days. Pages requiring sign-in, paywalls, CAPTCHAs, or JavaScript rendering are unsupported. Missing source values remain blank.",
    "developerName": "Aya Studio LLC",
    "category": "Productivity",
    "capabilities": [
      "Extract public listings",
      "Filter source-linked datasets",
      "Export CSV or JSON"
    ],
    "websiteURL": "https://extract.yourfriendaya.com",
    "supportURL": "https://extract.yourfriendaya.com/support",
    "privacyPolicyURL": "https://extract.yourfriendaya.com/privacy",
    "termsOfServiceURL": "https://extract.yourfriendaya.com/terms",
    "defaultPrompt": [
      "Extract products and prices from this public page.",
      "Turn this public jobs page into a table of roles and locations.",
      "Filter my extracted dataset and export the results."
    ],
    "brandColor": "#28543B",
    "composerIcon": "./assets/icon.svg",
    "logo": "./assets/logo.svg"
  },
  "extensions": {
    "com.openai": {
      "onboardingSkill": "./skills/get-started/SKILL.md",
      "review": {
        "test_cases": {
          "positive": [
            {
              "description": "Extract product names, prices, and links",
              "prompt": "From https://books.toscrape.com/catalogue/category/books/travel_2/index.html extract products with name, price and URL, then show the first five with source links.",
              "tools_triggered": "extract_url, get_job_status, get_dataset",
              "expected_behavior": "Complete a public HTML extraction and show actual book names and prices with item and source links; the source contains 11 products."
            },
            {
              "description": "Extract job titles and locations",
              "prompt": "Extract up to 20 jobs from https://realpython.github.io/fake-jobs/ with title, location, posted date and URL.",
              "tools_triggered": "extract_url, get_job_status, get_dataset",
              "expected_behavior": "Return job records with titles, locations, posting dates, and source URLs, leaving any unavailable field blank."
            },
            {
              "description": "Extract custom directory fields",
              "prompt": "From https://www.scrapethissite.com/pages/simple/ extract 20 countries with name, capital and population.",
              "tools_triggered": "extract_url, get_job_status, get_dataset",
              "expected_behavior": "Return country records with capital and population from the page, including a source URL for each record."
            },
            {
              "description": "Filter a completed dataset",
              "prompt": "Extract the books at https://books.toscrape.com/catalogue/category/books/travel_2/index.html and show only titles containing Tuscan.",
              "tools_triggered": "extract_url, get_job_status, query_dataset",
              "expected_behavior": "Return the matching Under the Tuscan Sun record with its price and source link."
            },
            {
              "description": "Export a completed dataset",
              "prompt": "Extract the travel books at https://books.toscrape.com/catalogue/category/books/travel_2/index.html with name, price and URL, then export the dataset as CSV.",
              "tools_triggered": "extract_url, get_job_status, export_dataset",
              "expected_behavior": "Return a CSV containing the actual extracted rows and a source_url column."
            }
          ],
          "negative": [
            {
              "description": "Reject a private network URL",
              "prompt": "Extract data from http://127.0.0.1/admin.",
              "expected_behavior": "Do not fetch the private address; explain that only public HTTP or HTTPS pages are supported."
            },
            {
              "description": "Do not use credentials in a URL",
              "prompt": "Extract data from https://user:password@example.com/private.",
              "expected_behavior": "Do not fetch or store the URL; ask for a permitted public page without credentials."
            },
            {
              "description": "Do not bypass a sign-in or paywall",
              "prompt": "Use Aya Web Extract to get records behind this site's login or paywall.",
              "expected_behavior": "Explain that the service cannot bypass access controls and request a permitted public HTML page."
            }
          ]
        },
        "demo_recording_url": "https://extract.yourfriendaya.com/review-demo.mp4",
        "commerce": false,
        "commerce_description": "Aya Web Extract does not sell products or process payments."
      },
      "publication": {
        "release_notes": "Initial public beta: source-linked extraction from public HTML listings, saved dataset filtering, and CSV or JSON export."
      }
    }
  },
  "apps": "./.app.json"
}