More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Project Indexer

    1

    **Your overrides** — a `path=note` line in `~project/.index-notes` (semicolon-free format: `scripts=one-line build helper` or a folder subtending with a trailing slash, e.g.

    $3.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Project Indexer

    Project Indexer

    Example session with this skill installed

    An agent starts a session in a folder it last saw a week ago. Instead of listing files and guessing, it reads INDEX.md: scripts/ — 'build + dev helpers', src/ — 'app entry points', docs/ — 'plans + runbooks'. It goes straight to src/main.ts.

    • Read your context and instructions
    • Compiled the project indexer

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

    About this skill

    Every time an agent (or you) opens a folder it hasn't seen in a week, it pays a re-exploration tax: list files, open a few, guess which one matters. For agent sessions that tax is tokens; for humans it is context-switching. This skill pays the tax once and writes the answer to disk: INDEX.md.

    First hit wins:

    1. Your overrides — a path=note line in ~project/.index-notes (semicolon-free format: scripts=one-line build helper or a folder subtending with a trailing slash, e.g. src/=app entry points).
    2. The folder speaks — the first meaningful line (comments stripped) of that folder's AGENTS.md or README.md.
    3. The truth of the count — "7 files", when nothing else is available.

    The result is honest by construction: it is derived from what exists, and your two-second override (echo 'ops=runbook + backup jobs' >> .index-notes) beats anything clever.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.

    Notes are derived from what exists: a badly written README first line will show up as that folder's note (fix the README or override in .index-notes). One line per folder by design (~80 chars) keeps the whole page under ~2K tokens. Regenerate after structural changes — INDEX.md is fully rebuilt each run.

    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 buying a skill.

    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 15 days ago

    Software
    BashGNU coreutils
    Optional
    tree

    Linux has these already; on macOS run brew install coreutils for realpath. tree draws the overview block, with find as the fallback.

    Permissions

    Read Files
    Write Files

    File Scopes

    writes
    ONE
    file:
    INDEX.md
    at
    the
    project
    root
    you
    point
    it
    reads
    project's
    folder
    structure
    .index-notes
    overrides
    and
    README/AGENTS.md
    first
    lines

    Purpose: generate a one-page folder index (INDEX.md) for any project. The script uses only its own bash + standard coreutils (realpath, find, grep, wc, date) plus optional tree. It never talks to the network, never reads environment variables, never runs anything from your project — it derives notes from plain text files and writes exactly one Markdown file.

    Listed15 days ago
    Updated7 days ago

    What's inside

    Frequently Asked Questions

    Because the honest map is small. One 80-character line per top-level folder answers the routing question — what is this folder for, where do I go — and anything longer stops being read by busy agents.

    First hit wins: your .index-notes overrides (path=note lines), then the folder's own AGENTS.md/README.md first meaningful line, then a bare file count. Honest by construction: derived from what exists, overridable in seconds.

    No — it routes, not searches. Pair it with folder search utilities (e.g. Ask Your Corpus): the map says where to look; the search digs into the folder.

    It reads structure and text-file first lines, writes one Markdown file, and runs nothing from your project. No network, no environment reads, no credentials.

    New in LLM Ops