Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Rss Api Watcher

    1

    Watch RSS feeds, JSON APIs, and GitHub endpoints with watermark deduplication: a state file records last-seen item IDs per source, so each poll emits only newco

    Free

    2 usesSecurity scanned
    Rss Api Watcher

    Rss Api Watcher

    Example session with this skill installed

    Initialize the watcher for my sources in watcher.yaml with a backfill of 20 items, then run a poll and send any new items to my webhook at http://api.internal/hooks/updates.

    • Read your context and instructions
    • Compiled the rss api watcher

    $ watcher init --state ./state.json --backfill 20
    State initialized. 20 items marked as seen.

    $ watcher poll --notify webhook=http://api.internal/hooks/updates
    Fetching 2 sources...
    releases: 0 new items
    news: 2 new items
    Webhook sent for 2 items. State updated.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    The problem

    Polling RSS feeds or JSON APIs often results in duplicate notifications or redundant data processing when a service restarts or feeds reorder items. Developers struggle to maintain state between polling intervals without building custom database logic.

    What it does

    • Tracks last-seen item identifiers using a local state file to prevent duplicate processing.
    • Supports RSS feeds, GitHub release APIs, and generic JSON endpoints.
    • Performs atomic watermark updates to ensure data integrity even if a process crashes.
    • Generates structured Markdown reports summarizing changes over a specific timeframe.
    • Triggers webhooks or email notifications only for genuinely new entries.

    Frameworks & tools

    Python, YAML configuration, JSON pointers, and standard webhooks.

    Why this beats prompting it yourself

    Writing a reliable deduplication engine requires handling atomic file writes, ID extraction logic, and state management. This skill provides a tested, CLI-driven workflow that handles the edge cases of backfilling and state resets out of the box.

    Use cases

    • Monitor GitHub releases for multiple dependencies to trigger CI/CD pipelines.
    • Aggregate technical blog feeds into a daily Markdown summary for team review.
    • Send Slack alerts via webhooks for specific API changes without spamming the channel.
    • Audit historical changes across various sources using the structured reporting tool.

    Known limitations

    Not suitable for real-time push requirements or sources lacking stable item IDs. Polling frequency should remain above 60 seconds to avoid rate limiting.

    Requires sources with stable item IDs. Not for real-time push needs. Polling must be spaced at least 60 seconds apart to prevent rate limiting.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    2 uses

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 3 months ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed3 months ago

    What's inside

    Frequently Asked Questions

    The skill uses a local state file to track unique item IDs (like RSS GUIDs or GitHub release IDs). It only triggers alerts for IDs it has never seen before, preventing duplicates even if a feed is reordered.

    Yes. Use the init command with the --backfill flag to mark the most recent entries as seen. This ensures your first poll only captures genuinely new updates moving forward.

    It works with any standard RSS/Atom feed, GitHub Release APIs, and custom JSON endpoints where you can define an ID extractor using JSON pointers.

    The tool performs atomic write-then-rename operations on the state file. If a process crashes, the watermark remains at the last successful poll's state to prevent data loss.

    New in DevOps & Cloud