In-session budget setting
Set a session budget from inside Claude Code using commands such as /redline 10m $5, /redline 1h 200k, or /redline 45m 10%.
Redline
Redline is a budgeting tool for Claude Code that paces sessions to stay within time, token, cost, or plan-percentage limits. It uses Claude Code’s native hooks and statusline to help sessions finish with a usable result instead of stopping abruptly.
Overview
Redline is a budgeting tool for Claude Code that lets you set limits for time, token usage, cost, or plan percentage inside an active session. It uses Claude Code’s own hooks and statusline so the session can pace itself toward the limit instead of ending abruptly at the ceiling.
The project is designed to keep work usable at the end of the budget: it aims to finish with a complete result, uses a reserve zone to block new tool calls before the hard limit, and counts subagent activity under the same session budget. The repository describes Redline as working with Claude Code today, with Codex and other agents listed as roadmap items.
Core capabilities
Live statusline feedback
Show a live burn-down in the statusline so you can see remaining time, cost, or plan headroom while the session runs.
Reserve-based pacing and enforcement
Pace the agent toward the budget instead of waiting for a hard stop, with a reserve zone that blocks new tool calls before the ceiling is reached.
Session-wide budget coverage
Treat the whole Claude Code session as one budget, including spawned subagents, so users do not need separate per-subagent limits.
Local control commands
Provide local commands to inspect active budgets, check landed-rate, view version info, or remove the integration.
Multiple install and removal paths
Install through Homebrew, a curl script, or a git clone workflow, then remove it later with redline uninstall.
Common use cases
Budgeting an active coding session
Set a limit for a Claude Code session when you want the model to pace work toward a fixed amount of time, spend, or tokens.
Leaving room for a usable finish
Keep longer tasks from ending abruptly by reserving enough budget for a wrap-up and final response.
Covering fanned subagents with one budget
Manage sessions that spawn subagents without having to assign separate budgets to each child task.
Inspecting budget performance locally
Monitor several active sessions and check whether they landed within budget using the local pulse command.
Adding and removing the integration
Install the tool into a Claude Code setup and remove it later if you no longer want the statusline and hook integration.
Pros and Cons
Pros
- Supports several budget types, including time, dollar cost, token count, and plan percentage.
- Uses Claude Code’s native hooks and statusline instead of a separate wrapper or daemon.
- Provides a live burn-down view so users can see budget headroom during the session.
- Applies one budget across the whole session, including fanned subagents.
- Offers multiple installation paths and can be uninstalled later.
Cons
- Wall-clock time enforcement is best-effort rather than precise; the architecture notes that time can pass between tool attempts or turn boundaries.
- The README says Redline works with Claude Code today, while Codex and other agents are still on the roadmap.
- The project depends on the Node runtime that ships with Claude Code, so setup is tied to that environment.
FAQ
What does Redline work with?
Redline is built for Claude Code and uses its native hooks and statusline rather than a separate process wrapper or daemon.
How do you install it?
The README shows installation options for Homebrew, a curl-based install script, or cloning the repository and running install.sh.
How do you set or clear a budget?
You can set budgets with commands like /redline 10m $5, /redline 1h 200k, /redline 45m 10%, or /redline off to clear the budget.
Are all budget dimensions enforced the same way?
The architecture doc says token, cost, and plan-percentage limits are measured precisely, while wall-clock time is best-effort because it is enforced at tool attempts or turn boundaries.
Quick Facts
- Category
- Developer Tool
- Primary use
- Token and time budgeting for Claude Code
- Platform
- Claude Code
- Runtime
- Node runtime that ships with Claude Code
- Source domain
- github.com
- License
- MIT
Альтернативы Redline
Macuse
Macuse is a macOS app that lets AI assistants control native Mac apps through MCP and use Computer Use for any app. It works with clients like Claude Desktop, Cursor, and Raycast, and keeps automation on-device.
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
Goldfish
Goldfish is an AI memory app for macOS that helps you reply, write, summarize, and continue using the context already on your Mac. It stores memory locally and writes in your tone.
Biji
Biji - это универсальная платформа, разработанная для повышения продуктивности с помощью инновационных инструментов и функций.
hob
hob is an independent workspace for coding agents that keeps agent sessions, terminals, history, and follow-up work organized around the tools and providers you already use. It is aimed at developers who want local control over routing, history, and workspace structure rather than a bundled model stack.
Claude Overlay
Claude Overlay is a Windows desktop overlay for Claude Code that reads your screen so you can ask questions, inspect content, and request edits without leaving the current app. It runs on your existing Claude subscription through the Claude CLI, so no separate API key is required.