repo nav
Generate hierarchical, progressive-disclosure Markdown indexes for codebase navigation by AI coding agents.
Free
Works with the AI Agents you already use
repo nav
Example session with this skill installed
Generate a baseline repo-nav index for this project. Use the current directory as the Designated Root Folder.
- Read your context and instructions
- Compiled the repo nav
I have initiated the repository navigation indexing for the current directory (Designated Root Folder: root). I used filelist.sh to generate the authoritative inventory and have processed the folder hierarchy from deepest to shallowest.
Repository Indexing Result
| File/Folder | Status | Summary |
|---|---|---|
DOCMAP.md | Created | Root index providing architectural overview and navigation guidance. |
src/auth/docmap.md | Created | Index for authentication logic, covering JWT and OAuth2 flows. |
src/db/docmap.md | Created | Index for database models and migration scripts. |
SCHEMA_MAP.md | Created | Map of all tables, views, and indexes defined in the DRF. |
ARCHITECTURE_MAP.md | Created | Overview of system layers and major responsibilities. |
# Designated Root Folder: Project Root
## Overview
This repository implements a distributed microservices architecture for order processing. It handles authentication, inventory management, and payment gateway integration.
## Navigation Guidance
AI agents should start with this `DOCMAP.md` to identify the relevant module. Folder-level `docmap.md` files provide specific file responsibilities and TODO markers.
## Primary Modules
- `src/auth/`: Authentication service (JWT, OAuth2)
- `src/orders/`: Core business logic for order lifecycles
- `src/common/`: Shared utilities and middleware (Merged into parent)
## Database Schema Summary
- **Table: Users** (Defined in `src/db/models/User.py`)
- **Table: Orders** (Defined in `src/db/models/Order.py`)
Next steps
- Review the generated
DOCMAP.mdat the project root. - Verify that
AGENTS.mdhas been updated with the new navigation instructions. - Run an incremental update after your next major commit to refresh the summaries.
Connects securely to your tools. The creator never sees your data.
About this skill
Large repositories often become a black box for AI coding agents, leading to high token costs and poor decision-making. This skill solves the "context window" problem by generating a hierarchical structure of index files using progressive disclosure. It builds a map of your codebase that allows agents to navigate directly to relevant files without scanning the entire tree.
What it does
- Hierarchical Indexing generates
DOCMAP.mdfiles at the root and folder levels to summarize architecture and business logic. - Progressive Disclosure ensures detailed information only appears at deeper levels, keeping agent context windows clean.
- Change Detection identifies modified files using size-based comparison for fast, incremental index updates.
- Cross-Cutting Maps creates specialized navigation for features, architecture, technology stacks, and database schemas.
- Smart Merging absorbs small folder summaries into parent indexes to reduce file clutter while preserving context.
How it works
- Plan & Confirm identifies the Designated Root Folder (DRF) and gets user approval for the indexing operation.
- Inventory Discovery uses
ripgrepvia mandatory shell scripts to create an authoritative file list. - Progressive Analysis processes folders from deepest to shallowest, generating LLM-backed summaries and semantic tags.
- Map Generation compiles specialized indexes like
SCHEMA_MAP.mdandFEATURE_MAP.mdfrom the docmap data.
Frameworks & tools
This skill requires ripgrep (rg) for repository discovery. It uses PowerShell on Windows and Bash on Unix-like systems for all automation and path handling.
Why this beats prompting it yourself
Manual prompts often result in hallucinated file structures or missing context in deep hierarchies. This tool enforces a strict reference workflow using shell-side validation and consistent templates, ensuring every file summary is evidence-based and every link is relative and functional.
Use cases
- Orienting a new AI agent to a massive legacy codebase.
- Generating a database schema map that connects SQL definitions to application code.
- Reducing token usage by preventing agents from recursively reading every file.
- Creating a feature map to track requirements against implementation and tests.
Known limitations
Does not support external systems like Jira or GitHub. It does not perform AST parsing, call graph analysis, or dependency extraction. Requires ripgrep to be installed and available on the path.
Requires ripgrep (rg). No AST parsing or call graph generation. Dependency extraction is reserved for future versions. No support for external systems like Jira.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 12 days ago
- Free with an Agensi account
Requires