Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    ELYRAON Secret & Environment Guard

    1

    Find exposed secrets and unsafe environment configuration before they reach source control or an AI agent.

    $9.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    About this skill

    Hardcoded credentials and careless environment handling are easy to commit and easy to hand to an AI agent by accident. ELYRAON Secret & Environment Guard is a local, offline scanner that reports likely secrets and unsafe environment-variable handling in the files you point it at, so you can review them before committing, sharing or giving a repository to another tool.

    It runs a local Python helper over UTF-8 source and configuration files, statically parses them, and returns a structured JSON report. It does not execute your code and does not change your files. Reported values are redacted and files are referenced by opaque IDs, so the report itself is not a new place for secrets to leak.

    What it does

    • Secret detection flags likely hardcoded credentials, including high-entropy strings and known credential formats.
    • Environment handling checks flag patterns where environment variables are handled unsafely.
    • Contextual heuristics help separate likely real secrets from test fixtures and placeholders.
    • Redacted, structured output gives schema-validated JSON findings you or your agent can review.

    What it is not

    It is a detector and reporter, not an enforcer. It does not revoke or rotate credentials, fix files, validate whether a found secret is still active, or replace a secrets manager. Detection is heuristic, so expect false positives and misses, and treat an empty report as "nothing found", not "safe". See Known Limitations.

    Static, heuristic scan of UTF-8 source and configuration files; the scanned code is not executed. Scans files present under the selected root; Git commit history is not inspected. Binary files, multiline keys and files over 1 MiB are not supported, and the maximum scan is 2000 files. Detection can produce false positives and can miss secrets, so an empty report does not prove a repository is safe and some findings need human review. It cannot tell whether a discovered secret is currently active. Offline only; it reports findings and does not revoke, rotate or remove credentials or modify files. Requires Python 3.11 or newer and Linux or macOS with POSIX descriptor-relative file operations. Tested on Linux with Python 3.12; other supported environments are unverified.

    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

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    Requires

    Python
    macOS

    Permissions

    Terminal / Shell
    Read Files

    Reads the local files and directories you point it at and runs a local Python helper that statically scans them and prints a JSON report. It does not execute the scanned code and does not modify target files. It makes no network requests and uses no external hosts. It does not require your environment variables or any credentials to operate. Files it inspects (including .env files) and the secrets it may find are separate from the permissions the guard itself needs.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    UTF-8 source and configuration files in a local directory or the specific files you point it at. It looks for likely hardcoded secrets (including high-entropy strings and known credential formats) and unsafe environment-variable handling, and returns a structured JSON report with redacted values.

    No to all three. It statically scans files with a local Python helper, does not execute the scanned code, does not modify your files, and works offline. It does not need your environment variables or any credentials to run.

    No. Detection is heuristic, so false positives and missed secrets are possible, and it cannot validate whether a found secret is currently active. It reports findings only; it does not revoke or rotate credentials. If it reports a secret, review the finding yourself and rotate the credential through its provider if it is real. Limits: binary files, multiline keys and files over 1 MiB are not supported, with a maximum of 2000 files.

    Review the finding yourself, since detection is heuristic and may be a false positive. If the credential is real, revoke or rotate it with its provider and replace it with a reference to configuration. The guard only reports findings; it does not rotate, revoke or remove anything.

    Findings use redacted values and opaque file IDs, so the JSON report is designed not to expose the actual secret values it found.

    New in Security