index

Generate indexed component and external specs with content-hash idempotency.

4|Updated Apr 26, 2026
One-click install
npx skills add https://github.com/curdx/curdx-flow --skill index-curdx
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: index
Source: https://github.com/curdx/curdx-flow/tree/main/plugins/curdx-flow/skills/index
Command: npx skills add https://github.com/curdx/curdx-flow --skill index-curdx

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires node, git, and includes scripts (resource) components.

What problem does it solve?

It reduces context rot and missing coverage by turning a repository’s code and relevant external resources into structured, discoverable component and external specs you can reuse across future deliveries.

Core Features & Use Cases

  • Repo component indexing: Scans controllers/services/models/helpers/migrations (configurable via arguments) and generates component spec files under specs/.index/components/.
  • Topology + routing guardrails: Builds a project topology snapshot to decide what to index and can halt early when required roots are missing.
  • External documentation & tool capture: Fetches and documents external URLs and also documents MCP resources or other plugin skills into specs/.index/external/.
  • Change-aware and safe re-runs: Supports --dry-run, --changed, and content-hash skipping to avoid unnecessary rewrites.

Quick Start

Run the indexing workflow for your repo by calling /curdx-flow:index with the options you want, such as limiting the scan to src and skipping deep interviews.

Frequently Asked Questions about index

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate structured documentation for my monorepo codebase components?

Codebase component indexing scans controllers, services, models, and helpers to generate structured component spec files under specs/.index/components/ for future delivery workflows. It builds a topology snapshot to decide what to index and halts early when required roots are missing.

What is the best way to capture external URLs and MCP documentation into reusable specs?

External documentation capture fetches external URLs and documents MCP resources or plugin skills into specs/.index/external/ spec files. It applies deterministic hashing and robust error handling to ensure recoverable fetch failures do not corrupt the generated index.

Can I use git-changed incremental scanning to update only modified component specs?

Yes, change-aware re-runs support the --changed flag to perform incremental scans based on git-changed files. Combined with content-hash skipping, this avoids unnecessary rewrites and ensures idempotent spec generation across safe re-runs.

Does the codebase indexing workflow require Node and Git as dependencies?

Yes, this indexing workflow requires both Node and Git as dependencies. Node executes the scripts for topology snapshot generation and spec writing, while Git enables incremental scanning via git-changed files and supports change-aware re-runs.

How do I preview codebase mapping results without writing spec files to disk?

You can use the --dry-run flag to preview codebase mapping results without writing any spec files to disk. This allows you to validate topology snapshots and scan coverage planning before committing to actual spec generation.

Why does codebase indexing halt early when required project roots are missing?

Codebase indexing halts early when required project roots are missing to enforce topology and routing guardrails. This prevents incomplete or inaccurate component spec generation by ensuring the repository structure meets the minimum scanning requirements before processing.