Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    logscry

    1

    It never sends logs to a remote model without asking you first, and *never puts an API key on the command line*.

    Free

    1 useSecurity scanned

    About this skill

    The skill teaches an agent to pick the right source, bound live follows with timeout, read logscry’s JSONL export instead of the screen, and report anomalies by pattern rather than pasting raw lines. It never sends logs to a remote model without asking you first, and never puts an API key on the command line.

    Requires the logscry binary (free, Apache-2.0, Linux/macOS/Windows): https://github.com/maxie7/logscry Source of this skill, always the latest version: https://github.com/maxie7/logscry/tree/main/skills/logscry

    Burst detection needs 1m+ history. Stdin pipes disable burst logic. Level detection requires specific formats (e.g., [ERROR] or level=error). Best-effort anonymization is not a guarantee.

    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

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 26 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed26 days ago

    What's inside

    Frequently Asked Questions

    Logscry uses a specialized scoring engine to filter out routine noise and mask sensitive data before it reaches the AI. This saves you significant token costs and prevents the "context window fatigue" that happens when you feed raw logs directly into a generic prompt.

    The skill is highly versatile, supporting live Docker container streams, systemd journal logs, piped command outputs (stdin), and static log files. It also supports standard JSON log formats.

    Logscry is built as a Go binary and is compatible with any agent or workflow that can execute CLI commands. It integrates with OpenAI-compatible APIs and local LLM providers like Ollama for autonomous explanations.

    The skill includes a 'Sensitive Data Masking' feature that identifies and anonymizes IPs, emails, and potential secrets. This ensures that high-novelty anomalies are triaged while keeping your private data out of the model's training set or logs.

    You will receive the logscry binary and documentation on how to integrate the scoring engine into your existing observability stack. Updates are released periodically to improve templating accuracy and add support for new log formats.

    While it performs real-time analysis, burst detection typically requires a one-minute warm-up period of log history to establish a baseline for what constitutes a "normal" frequency.

    New in DevOps & Cloud

    Free