Repo Prompt

The Context IDE for AI Agents

Automated context curation, deep analysis, and multi-agent orchestration from a single native app.

Download for Free

From prompt to verified delivery

Plan with a powerful reasoning model, then dispatch sub-agents across providers to implement, each scoped to its own files.

Prompt

Describe the task

Context Builder

Curates relevant code

Discovery agent

Orchestrator decomposes plan into tasks Each sub-agent gets scoped file boundaries

Sequential

Deep reasoning model

  1. Implement auth service
  2. Add API validation
  3. Wire up middleware

Orchestrator steers between tasks

Plan.md shared source of truth

Parallel

OpenCodeExplore codebase

CodexBackend logic

GeminiUpdate configs

Multi-wait: unblocks when first completes

Orchestrator verifies adherence to plan then continues until complete

A Native Home for Your Coding Agents

Run CLI agents with full codebase context, token-efficient MCP tools, and built-in orchestration, all inside a beautiful Mac-native interface.

Codex CLI

App Server

Launch from Your Personal Agent

OpenClaw and other personal agents can launch Repo Prompt's orchestrate workflow via CLI, planning, decomposing, and dispatching sub-agents to complete complex engineering work while you sleep.

Context Builder Workflows

Included slash commands you can run from any MCP client

01

Explore

Agent navigates codebase

02

Curate

Select within token budget

Curated Context

  • auth.tsL5-89
  • api.tsL22-58
  • useAuth.tsL1-41
  • Header.tsxL12-67
  • Sidebar.tsxL1-34

Orchestrate

Plan, decompose into tasks, and dispatch sub-agents across providers to complete complex work autonomously

Code Reviews

Publishes git diffs alongside codebase context for reviews that understand what changed

Refactor Code

Two-pass: analyze for opportunities, then plan implementation preserving behavior

Investigate Issues

Systematic exploration with evidence gathering until root cause is found

Stop prompting models directly — orchestrate with Repo Prompt

Repo Bench

Measure AI Coding Performance

Don't guess which model is best for your codebase. Repo Bench provides data-driven rankings measuring large context reasoning, file editing precision, and instruction adherence.

100sModels Tested

CommunityDriven Results

RealtimeUpdates

Works With Every Agent

15+ token-efficient MCP tools, automated context curation, and multi-agent orchestration. Accessible from any MCP-compatible client.

Automated Context Curation

Context Builder explores your codebase, selects the most relevant files within a token budget, and hands curated context to powerful analysis models.

Multi-Agent Orchestration

Launch parallel or sequential agent sessions, steer them mid-flight, and let an orchestrator verify the work.

Codemaps & Token Efficiency

Function signatures via tree-sitter let agents see 10x more files. Slices, codemaps, and full content: the right detail level for every file.

Oracle Collaboration

Your agent consults powerful reasoning models mid-session. Context Builder finds the answer using your full repo, no manual file selection needed.

Repo Prompt Pricing

Free Tier

$0

Powerful features, zero cost

Key features include:

  • Mac-Native UX: Effortlessly build prompts from your codebase with a seamless interface.
  • Clipboard & Chat: Copy prompts or edit files via API chat (up to 32k tokens) for smooth workflow integration.
  • Workspaces: Open multiple folders or repos simultaneously (e.g., code + packages or docs) for comprehensive context.
  • Advanced Search & Filtering: Quickly locate files using regex and git-ignore filters for precise control.
  • Stored Prompts: Save and organize prompts for efficient reuse across your projects.
  • Apply Mode: Use ChatGPT or Claude outputs to instantly apply multi-file edits (create, modify, delete files; rename coming soon).
  • Model Access: Connect to local and hosted models (OpenAI, Anthropic, Gemini, DeepSeek, Azure, OpenRouter) - bring your own API key for maximum flexibility.
  • Claude Code & Codex CLI Integration: Connect Claude Code & Codex CLI to use your existing subscriptions in the built-in Chat.
  • Custom OpenAI-compatible providers: Add any supported provider, including LiteLLM.
  • Advanced model customization: Tune temperature, context windows, and per-model capabilities.

Pro Upgrade Options

Monthly subscription

$14.99/month

Flexible billing

1-week free trial to start

Includes all updates during subscription

Cancel anytime

Yearly subscription

$149.99/year

Best value

2 months free compared to monthly

Team Licenses

From $14.99/mo

Per seat pricing

Managed seats with centralized billing

Special Pricing Options

🎓 Student Discount

Are you a student? We offer special academic pricing! Email us from your .edu address to request a discount code.

For special pricing inquiries, contact: contact@repoprompt.com

Latest from the Blog

[Nov 04, 2025

**Context Over Convenience** \ Coding agents are convenient but wasteful with context. Planning with the full picture beats navigating in the fog.\ Read Article →](/content/blog/context-over-convenience/index.html)

[Oct 08, 2025

**Building Repo Bench** \ How I built a benchmark to measure what truly matters for AI coding assistants - file editing precision, instruction following, and handling complex context.\ Read Article →](/content/blog/building-repo-bench/index.html)