skills.rest
kapilvirenahuja
kapilvirenahujaCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

scan-codebase

Scan repository roots to generate scan-index.json and scan-result.yaml artifacts.

3|Updated Jan 25, 2026
Search Tags:#code generation#monorepo#STM#codebase scanning#repository intelligence#deterministic analysis#structured inputs
One-click install
npx skills add https://github.com/kapilvirenahuja/garura --skill scan-codebase
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: scan-codebase
Source: https://github.com/kapilvirenahuja/garura/tree/main/core/components/skills/scan-codebase
Command: npx skills add https://github.com/kapilvirenahuja/garura --skill scan-codebase

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It removes the guesswork and inconsistency of manually inspecting a repository by producing a deterministic, structured scan of the codebase for downstream reasoning.

Core Features & Use Cases

  • Deterministic brownfield scanning: Harvests repository structure and signals such as manifests, framework idioms, and git context into a bounded index.
  • STM-first structured inputs: Writes scan-index.json into STM as the sole structured-input source for every subsequent infer-from-code step.
  • Budgeted, tool-aware execution: Enforces size and time limits while conservatively omitting unavailable sections instead of failing the whole scan.
  • Use case: Bootstrap enterprise-grade code generation for an existing product by extracting tech stack, configs, and ADRs before any LLM inference runs in the /codify play.

Quick Start

Run the scan-codebase skill with repo_roots and an output_path to generate scan-index.json in STM for your codify workflow.

Frequently Asked Questions about scan-codebase

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

FAQPage Schema
How do I extract structured facts from a repository for code generation?▼

Scanning a codebase deterministically extracts structured facts like manifests, framework idioms, and git context into a bounded index. This provides downstream reasoning agents with evidence for code generation instead of directly browsing source code.

What is the best way to bootstrap enterprise code generation for an existing product?▼

Bootstrapping code generation for an existing product requires scanning the repository to extract tech stack, configs, and ADRs into a scan-index.json file. This provides deterministic evidence for downstream inference agents before any LLM inference runs.

Do I need absolute paths to scan a monorepo for repository intelligence?▼

Yes, repository scanning requires absolute repo roots and a writable STM output path. It enforces configurable size and time budgets while scanning one or more monorepo roots to generate the structured index.

Does codebase scanning fail when parts of the repository are unavailable?▼

No, codebase scanning conservatively omits unavailable sections instead of failing the whole scan. It enforces size and time limits to ensure budgeted, tool-aware execution while emitting a scan-result.yaml status artifact.

How does deterministic codebase analysis handle downstream reasoning?▼

Deterministic codebase analysis writes scan-index.json into STM as the sole structured-input source. This ensures every subsequent infer-from-code step reasons over deterministic evidence rather than directly browsing source code.

Related Skills

Explore More Software Engineering Skills→
kapilvirenahujakapilvirenahuja
3

infer-scope-from-code

Infer v1 scope.yaml by scoring KB capabilities against scan-index evidence.

Community
Advanced
kapilvirenahujakapilvirenahuja
3

infer-project-profile-from-code

Infer a project-profile artifact from scan-index.json evidence.

Community
Advanced
kapilvirenahujakapilvirenahuja
3

garura:codify

Infer Garura product LTM artifact proposals from codebases via deterministic scanning.

Community
Advanced
lmanchulmanchu
5

gstack

Controls a headless Chromium browser via CLI to navigate, extract, and interact with web pages.

Community
Advanced