Caveman 3.2.0 released
A release adding cache-stable compression, native install improvements, and support for new provider integrations across platforms.
What changed in 3.2.0
- Compression decisions now persist within conversations; rate-limit responses with Retry-After headers are passed through correctly.
- Native Claude and Codex installations include three voice skills; disabling native routing keeps skills available while turning off configured Claude profiles.
- Subagents now adopt their parent session's mode; headless Claude and Agent SDK sessions start with voice disabled.
- New provider support added for standalone OpenCode, OpenAI-compatible services, and integration extensions for Codex, Cursor, Copilot CLI, Antigravity CLI, Grok Build, and CodeBuddy.
- Browser extension updated to version 1.3.0 with improved rich-editor snapshots and send-control discovery.
Check before you update
- Standalone installer moved from bin/ to installer/ directory
Summarised by AgentGid from JuliusBrussee's official release notes on GitHub, as reported there; read the original before relying on it. Price, status and release dates come from our daily data.
Caveman at a glance
| What it is | Skill, local proxy and SDK that reduce the tokens coding agents write and read: the skill makes replies terse and the proxy compresses logs, JSON, test output and web pages before they reach the model. It works with Cla… |
|---|---|
| Latest version | 4.1.0, Nov 8, 2026. This page is about an earlier release. |
| Release pace | 5 stable releases in the last 30 days; 3.2.0 came the same day as the previous one |
| Price | Free + $300/mo. Caveman agent and SDK are free and open source (Apache-2.0). Caveman Cloud Free tier is $0/month for one person with usage allowances (100K events, 600 agent minutes, 25K routing … Official pricing ↗ (checked Oct 14, 2026) |
| Gid Score | 31 #41 in Agent tooling for cost, context and memory |