Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Plugin Status

    1

    One-screen health check for the jev-tools plugin: backend, key found or not, mode, and a summary of its log.

    Free

    1 useSecurity scanned
    Plugin Status

    Plugin Status

    Example session with this skill installed

    Check whether the jev-tools plugin is actually working. Use the status skill.

    • Read your context and instructions
    • Compiled the plugin status

    Real output of the packaged skill, run against a small hand-made sample log and a dummy key in an isolated home folder (no network, no real account). A fresh install with no key and no log prints "api key MISSING: every hook silently does nothing", then "no log yet", and exits 1.

    Python 3.13.15
    backend api (hosted api.codiv.ai)
    api key found via OPENJEV_API_KEY
    mode shadow (shadow = log only, active = block/inject, off = nothing sent)
    skill hook off (opt-in: it sends every prompt)
    log ~/.jev-tools/log.jsonl
    last 24h {'ask': 2, 'rule_check': 2, 'ask_error': 1}
    last ok 5 min ago, 580 ms
    last error ask_error HTTP Error 429

    Reading it: "ask_error HTTP Error 429" means quota or a rate limit. "rule_check" rows are normal in shadow mode.

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

    About this skill

    The jev-tools Claude Code plugin fails open: if the key is missing, Python is not on PATH, or the API is rejecting calls, its hooks quietly do nothing, and a broken install looks identical to a working one. This skill prints what is actually going on, on one screen.

    This is a companion to the jev-tools plugin (github.com/Rcidshacker/jev-tools, MIT). On its own, with no plugin installed, there is no decision log to read, and it will only tell you that.

    What it prints

    • Python version, and which backend is in use: hosted (api.codiv.ai), a local model, or offline
    • Whether an API key was found, and where from. Never the value.
    • The mode (shadow, active or off) and whether the optional skill hook is on
    • The decision log path, then a summary: events in the last 24 hours, minutes since the last successful model call, and the last error

    Exit code 1 means there is no log yet.

    Good to know

    • It only reads. It makes no network calls, sends nothing anywhere, and does not create the log if it does not exist.
    • It reads the log at ~/.jev-tools/log.jsonl (or the path in JEV_LOG), the optional ~/.jev-tools/config.json and credentials files, and environment variable names such as OPENJEV_API_KEY.
    • Python 3.10+, standard library only, nothing to install.
    • Tested on Claude Code on Windows. It follows the SKILL.md standard but I have not run it elsewhere.
    • Not a general-purpose health check. It checks the jev-tools plugin and nothing else.

    Requires Python 3.10+ on PATH and the jev-tools plugin. Does not create the log; without the plugin it prints "no log yet" and exits 1. Tested on Claude Code on Windows only.

    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 1 month ago

    • Free with an Agensi account

    Requires

    Python

    Permissions

    Terminal / Shell
    Environment Variables
    Read Files

    File Scopes

    ~/.jev-tools/**

    Shell runs one bundled Python script. Read files: the plugin log and optional config under ~/.jev-tools/, plus the skill folder. Environment variables: checks which key variable names are set, never prints values. No network calls and nothing is written.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    Yes, for it to be useful. This skill reads the decision log the plugin writes. With no plugin installed there is no log, and it will only say "no log yet" and exit 1.

    No. It reports only whether a key was found and where from (for example an environment variable name), never the value.

    No. It makes no network calls and sends nothing anywhere. It only reads local files and environment variable names.

    The log at ~/.jev-tools/log.jsonl (or the path in JEV_LOG), the optional ~/.jev-tools/config.json and credentials files, and environment variable names such as OPENJEV_API_KEY. It does not create the log if it is missing.

    The last-error line shows the most recent failure from the last 24 hours. HTTP Error 429 usually means quota or a rate limit; HTTP Error 403 usually means the key or a proxy. It reports what the log recorded; it does not test the key itself.

    Python 3.10+ on PATH, standard library only, nothing to install. I have tested it on Claude Code on Windows only.

    New in LLM Ops