AI Developer Tools

claude-thermos

An open-source utility that keeps your Claude Code prompt cache warm to avoid costly re-encodings when subagents run longer than five minutes.

claude-thermos logo

claude-thermos

Visit website

What is claude-thermos?

claude-thermos is an open-source tool that runs as a local reverse proxy to monitor Claude Code sessions and automatically sends minimal warm requests to prevent the prompt cache from expiring during subagent idle periods.

claude-thermos vs Similar AI Tools

Pricing ModelFreeFreeFreeFree
Free Credits
Key Features
  • Automatic cache warming during subagent idle periods
  • Adjustable idle time, interval, and max cycles
  • Event logging and session summaries with cost savings
  • Seven breakable boxes covering OWASP Agentic Top-10 vulnerabilities
  • Three guided simulations for cascading failures, human-agent trust, and rogue agents
  • Network-isolated Docker containers for safe execution
  • Code-to-runtime reasoning across cloud, Git, and Kubernetes
  • Action-gate enforces read-only policy on every API call
  • Sandboxed JavaScript execution for concurrent research
  • Append-only, SHA-256-addressed event history through Jaybase
  • AES-256-GCM encryption for stored node payloads
  • Unified RBAC for ledger, notes, snapshots, and audit reads
Pros
  • Reduces API costs by keeping cache warm
  • Easy to set up with uvx
  • Covers full OWASP Agentic Top-10 in a realistic manner
  • Docker isolation prevents accidental damage
  • Read-only by construction prevents accidental writes
  • Evidence-backed verification cross-checks every finding
  • Opinionated and secure accounting CLI with immutable audit trail
  • Designed for AI agent integration with JSON output
Cons
  • Requires Python 3.11+
  • Only works with Claude Code CLI
  • Requires Docker and technical setup
  • Not for production use; only for lab environments
  • Requires an LLM API key, incurring token costs
  • Limited to read-only operations, cannot remediate
  • Pre-1.0, limited feature set
  • No native QuickBooks import (agents must normalize data)
Best For
  • Developers using Claude Code for complex tasks with many subagents
  • Teams looking to optimize API spending on Claude Code sessions
  • Security researchers focusing on AI agent vulnerabilities
  • Developers building MCP-based applications
  • Security engineers
  • DevOps teams
  • Small teams needing secure, auditable accounting with AI agent support
  • Developers integrating automated bookkeeping workflows

How to use claude-thermos?

  1. 1Ensure Python 3.11+ and the 'claude' CLI are installed and on your PATH.
  2. 2Run Claude Code through claude-thermos using uvx: 'uvx claude-thermos' (instead of 'claude').
  3. 3Pass any Claude arguments as usual: 'uvx claude-thermos -p "fix the bug"'.
  4. 4Optionally tune behavior with flags: --idle, --interval, --max-cycles, --subagent-window.
  5. 5To disable warming for a run, set environment variable CLAUDE_WARMER_DISABLE=1.

claude-thermos Key Features

  • Automatic cache warming during subagent idle periods
  • Adjustable idle time, interval, and max cycles
  • Event logging and session summaries with cost savings
  • Seamless integration with existing Claude Code usage

claude-thermos Use Cases

  • Reducing costs in long Claude Code sessions with multiple subagents
  • Avoiding prompt cache expiry when the main agent is blocked on subagents
  • Monitoring and optimizing API usage for Claude Code deployments

claude-thermos Pricing & Free Credits

claude-thermos currently operates on a Free model.

This tool is completely free to use

Free

$0

Open-source tool, no cost to use.

claude-thermos Pros & Cons

Pros

  • Reduces API costs by keeping cache warm
  • Easy to set up with uvx
  • Provides detailed logs and savings estimates
  • Open-source and customizable

Cons

  • Requires Python 3.11+
  • Only works with Claude Code CLI
  • Additional local proxy setup may add complexity

What is claude-thermos best for?

  • Developers using Claude Code for complex tasks with many subagents
  • Teams looking to optimize API spending on Claude Code sessions

claude-thermos FAQ

Top free alternatives to claude-thermos

Openbase logo

A voice-controlled IDE that enables developers to initiate AI coding sessions with Codex or Claude Code, approve commands, and review diffs from their phone.

Free
SureWire logo

SureWire is a specialized QA platform that stress-tests AI agents for safety, reliability, and compliance using purpose-built testing agents.

Free
Notte logo

Browser infrastructure platform for AI agents to run on the internet at speed with cloud browser sessions, agents, and serverless functions.

Free
YAFL logo

An agent-first file transfer tool that enables secure, encrypted file sharing between AI agents via MCP calls without human involvement.

Free
Manifest logo

Manifest converts any URL into a structured JSON map of what AI agents can interact with on a page—buttons, forms, inputs, and required fields.

Free
B

Personal GitHub Pages site by Basert, currently displaying default welcome content and instructions for using GitHub Pages with Jekyll.

Free
Termaxa logo

A cooperative gate for shell commands that AI agents run, providing previews, backups, policy enforcement, and audit for tools like Claude Code and Cursor.

Free
Agentcard logo

Agentcard provides agent-friendly card issuing and payment infrastructure for AI agents, enabling 5-minute setup and autonomous purchases.

Free

Best alternatives AI Tools to claude-thermos

mcploitable logo

A collection of deliberately vulnerable MCP servers for training in agentic security, mapped to the OWASP Top 10 for Agentic Applications.

Cynative logo

Open-source AI tool for deep infrastructure research, running frontier models across code, cloud, and runtime to deliver verified answers.

Magpie logo

Magpie is an opinionated accounting CLI for humans and AI agents, providing double-entry bookkeeping with RBAC and immutable event storage on Jaybase.

agent-manager logo

Terminal UI to manage AI coding-agent sessions (Claude Code, OpenCode, Codex, Grok Build) in tmux with live status, group tree, and diff review.

Openbase logo

A voice-controlled IDE that enables developers to initiate AI coding sessions with Codex or Claude Code, approve commands, and review diffs from their phone.

Free
OpsCat logo

Zero-config, single-binary software catalog with auto-discovery, dependency visualization, compliance scorecards, and native MCP integration for AI agents.

BrowserAct Skills logo

Browser automation CLI for AI agents to bypass anti-bot walls, hand off to humans, and run parallel tasks.

lee-ai logo

An AI-powered shopper assistant that provides a live cursor to guide website visitors, point out products, and display price calculations.