Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.joggr.ai/llms.txt

Use this file to discover all available pages before exploring further.

Repo Context is everything that shapes how an agent behaves within your codebase — your coding standards, architecture, naming conventions, file structure, and rules. Without it, agents guess. With it, they write code that matches your team’s expectations from the first prompt.

​
What is Repo Context

Repo Context is three things:
  1. AI Instructions — the top-level files each agent reads at the start of every session: CLAUDE.md, AGENTS.md, .cursorrules, copilot-instructions.md. Joggr generates these automatically, tailored to each provider’s native format.
  2. AI Rules — granular rule files that enforce specific conventions: .claude/rules/, .cursor/rules/. More focused than instructions — one rule per file, composable and easy to maintain.
  3. Internal Docs — markdown documentation that lives in your repo: coding standards, architecture docs, READMEs, guides, and API references. jog docs generates and keeps these current so agents always have an accurate picture of your codebase.
Every agent session starts by reading these files. Instead of guessing, the agent knows your stack, your patterns, and your boundaries from the first prompt.

​
Why Repo Context matters

Without it, every agent is a new hire that hasn’t read the docs. You end up reviewing the same mistakes, correcting the same patterns, and re-explaining the same conventions — per session, per agent, per tool. Repo Context solves three problems:
  1. Consistency — every agent follows the same standards, regardless of which tool you use. Claude Code, Cursor, Copilot — same conventions, same output quality.
  2. Context — agents understand your architecture, dependencies, and patterns before they start. No cold starts, no guessing.
  3. Maintenance — Joggr keeps instruction files in sync with your codebase. jog drift detects when anything goes stale before it misleads an agent.

Agent-Agnostic

One command generates instruction files for every AI tool in your stack, each in its native format — CLAUDE.md, .cursorrules, AGENTS.md, copilot-instructions.md. Switch tools without rewriting configs.

Codebase-Aware

Joggr analyzes your code to infer standards, patterns, and architecture — not boilerplate templates.

Always Current

jog docs goes deep — generating full codebase documentation that stays current as your code evolves. jog drift catches anything that goes stale.

​
How it works

1

Initialize

jog init scans your codebase — tech stack, file structure, naming conventions, dependencies, and existing documentation. It builds a profile of how your team works.
2

Generate Instructions

From that profile, Joggr generates AI Instructions and AI Rules for each tool you’ve selected — placed where each agent expects them: repo root, .claude/, .cursor/, .github/.
3

Generate Docs

jog docs goes deep — generating coding standards, architecture docs, API references, and guides that agents can reference on every task.
4

Maintain

jog drift detects when anything goes stale before it misleads an agent.