Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Find Files

    1

    Rank the files in a repo by relevance to a plain-language question, so you open the right two or three first. Works with no API key.

    Free

    1 useSecurity scanned
    Find Files

    Find Files

    Example session with this skill installed

    Use find-files to find where login token refresh is handled in this repo. Show me the top 5 files.

    • Read your context and instructions
    • Compiled the find files

    Summary

    The most likely file is auth/token.py. It ranked first in both runs below, and the next two are worth a look.

    These are real outputs from the packaged skill on a small throwaway repo (7 files, a made-up shop-api). The query was "how is login token refresh handled". Paths are shortened.

    Run A: no key, keyword ranking only

      100  auth/token.py
       82  tests/test_token.py
       59  auth/session.py
       29  api/routes.py
       24  README.md
    

    The note on stderr says Jev unavailable: keyword order only. This needs no key and no network.

    Run B: with a stand-in model

    For this run I pointed the skill at a tiny local stand-in that marks files defining refresh_token as most relevant. It shows how the model path ranks and prints. It says nothing about how a real model performs.

      100  auth/token.py
       75  tests/test_token.py
       75  auth/session.py
       25  api/routes.py
       25  README.md
    

    What to open

    RankFileWhy it is there
    1auth/token.pydefines refresh_token
    2tests/test_token.pytests the refresh behaviour
    3auth/session.pycalls refresh_token on each touch
    4api/routes.pythe ping route that keeps a login alive
    5README.mdmentions refresh in prose only

    Next steps

    • Open auth/token.py first, then auth/session.py to see who calls it.
    • Treat the list as a hint. In the author's own test, keyword counting alone did about as well as the model.
    • For a model second opinion, set a free Codiv OpenJev key as OPENJEV_API_KEY and run it again.
    • Use --top 8 for more results, and --root <folder> to search somewhere else.

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

    About this skill

    Hey! This one helps you find your way around a codebase you don't know yet.

    Ask a question in plain words, like "how is login token refresh handled?", and it gives you a short ranked list of the files most likely to answer it, so you open the top two or three instead of searching blind.

    How it works

    1. A keyword walk of the repo collects up to 100 candidate files. It skips .git, node_modules, virtual environments, build folders, files over 300 KB, and files with secret-looking names like .env or *.pem.
    2. Optional model scoring. A small decision model (OpenJev) scores candidates from short excerpts, keeps the top 10, then re-judges those from longer excerpts. The final order blends both passes.
    3. No model? No problem. With no key you get the keyword order, with a note on stderr. It needs no key and no network.

    What you get back

      100  auth/token.py
       82  tests/test_token.py
       59  auth/session.py
       29  api/routes.py
       24  README.md
    

    One line per file, percent path, best first.

    Using the model, and what leaves your machine

    Model mode needs a free Codiv OpenJev API key as OPENJEV_API_KEY, or a small local model set up with jev-tools-setup. In model mode, short excerpts of the candidate files (their opening lines and the lines matching your keywords) are sent to api.codiv.ai, or to a loopback address for a local model. Secret-shaped strings are redacted first, and the key is only ever sent to those hosts. For code you would rather not share, just don't set a key. A full run over 100 candidates costs tens of thousands of input tokens.

    What you need

    Python 3.10+. Standard library only, nothing to install.

    Honest limits

    • It is a hint, not an oracle. On 8 labelled queries across 8 cloned repositories, plain keyword counting put the defining file in the top 3 for 4 of 8. The model ranking got 6 of 8 on one run and 4 of 8 on an identical rerun. So it was not measurably better than keywords, and repeat runs differ by up to 2 of 8. Open the top two or three files rather than trusting rank 1.
    • READMEs and docs can outrank the code that implements the thing. Skip them when you want code.
    • "No keyword candidates" means your words appear nowhere in the repo. Rephrase with identifiers likely to be in the code.
    • It appends one small line per model call (timestamp, number of questions, milliseconds, tokens, or an error type) to ~/.jev-tools/log.jsonl, or to the path in JEV_LOG. File contents and keys are never logged.

    Part of jev-tools

    This skill is one of several from the free, MIT-licensed jev-tools project. The full plugin adds optional hooks that this listing does not include: github.com/Rcidshacker/jev-tools. This skill works fine without it.

    Needs Python 3.10+. Searches text files up to 300 KB and skips .git, node_modules, virtual environments, build folders and secret-looking file names. Keyword-based, so it only considers files that contain at least one of your query words. Not measurably better than plain keyword counting in the author's 8-query test, and repeat runs of the model ranking can differ by up to 2 of 8. Not for exact-symbol lookups (use grep). Tested by the author 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

    Codiv
    Python

    Permissions

    Terminal / Shell
    Environment Variables
    Read Files
    Write Files
    Network Access

    Allowed Hosts

    api.codiv.ai

    File Scopes

    ~/.jev-tools/**

    Shell runs the bundled Python script. Files: it reads the files under the folder you point it at (to rank them), and the optional config and credentials files that jev-tools-setup writes in ~/.jev-tools. It appends one small log line per model call there (never file contents or keys). Environment: reads OPENJEV_API_KEY (and JEV_MODE, JEV_LOG) if you set them. Network: only in optional model mode, a POST to api.codiv.ai (or a loopback address for a local model), with secret-shaped strings redacted first; redirects are refused. With no key it makes no network calls.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    You ask a question about a codebase in plain words, and it prints a short ranked list of the files most likely to answer it, as "percent path", best first. It collects keyword candidates from the repo, and an optional model re-scores the best ones. It is a hint for where to look first, not an answer.

    No. Without a key it ranks by keyword counts and tells you so on stderr ("Jev unavailable: keyword order only"). If you want model scoring as well, set a free Codiv OpenJev API key as OPENJEV_API_KEY, or set up a small local model with jev-tools-setup (see the repo). The model was not measurably better than keywords in the author's test, so the key is optional.

    With no key, nothing. In model mode, short excerpts of up to 100 candidate files (the opening lines and the lines that match your keywords, up to 2,500 characters each) go to api.codiv.ai, or to a local server if you set one up. Files with secret-looking names such as .env or *.pem are never read, and secret-shaped strings are redacted before sending. That is a safety net, not a guarantee, so skip the key for code you would not want to share.

    Modestly. On 8 labelled queries across 8 cloned repositories, plain keyword counting put the file that defines the thing in the top 3 for 4 of 8. The model ranking got 6 of 8 on one run and 4 of 8 on an identical rerun. So treat the list as a hint, and open the top two or three files rather than trusting rank 1.

    It is a standard SKILL.md that runs a Python script, so any agent that follows the SKILL.md standard and can run Python should work. I built it for Claude Code, where the skill folder is available as ${CLAUDE_SKILL_DIR}, and I have not tested other agents. In another agent, point the command at wherever the skill folder was installed.

    New in Code Quality & Review