Show HN: Claude-thermos – keeps your Claude session warm for you

A new open-source tool called claude-thermos prevents expensive prompt-cache rebuilds in Claude Code sessions by keeping the main agent's cache warm while subagents run. The tool, which runs as a local reverse proxy, targets the 5-minute TTL on Anthropic's prompt cache that causes roughly 22% of total costs on long sessions with subagents. Developer 'nchammas' released claude-thermos on GitHub, claiming it eliminates the cost of re-encoding cached conversation history.

Stop paying to rebuild your Claude Code cache. When your main agent waits on a subagent for more than 5 minutes, its prompt cache silently expires, and the next turn re-encodes your entire conversation at the write rate instead of reading it back cheap. On long sessions with many subagents that's roughly 20% of your bill. claude-thermos keeps the cache warm so you never pay that tax. Run Claude Code exactly as you normally would, but through claude-thermos with uvx : uvx claude-thermos instead of: claude uvx claude-thermos -p "fix the bug" any claude args pass straight through Requires Python 3.11+ and the claude CLI on your PATH . That's it. Warming runs automatically in the background. To disable it for a run without changing the command, set CLAUDE WARMER DISABLE=1 . Tuning all optional : | Flag | Default | Meaning | |---|---|---| --idle | 270 | Seconds the main agent must be idle before warming kicks in | --interval | 270 | Seconds between warming cycles | --max-cycles | 4 | Max warms per idle episode auto for unlimited | --subagent-window | 540 | Seconds a subagent counts as "still active" | Claude Code's prompt cache uses a 5-minute TTL . Every turn, your whole conversation history is served from cache at 0.1x the input price instead of being re-sent at full price, as long as the cache stays alive. The cache expires if more than 5 minutes pass between requests on the same prefix. The dominant trigger for that gap is not you thinking. It's the main agent blocked on a subagent that runs longer than 5 minutes . A subagent has a different system prompt and tool set, so its requests have a different cache prefix and never refresh the main agent's. While the subagent works, the main agent's cached history ages untouched; past 5 minutes it's gone. When the subagent returns, the main agent resumes with a byte-identical, append-only history, and finds its cache missing, forcing a full re-encode at the 1.25x write rate. By then the history is large, so the re-encode is expensive: individual collapses re-write 200K to 500K tokens. Measured across roughly 185 local sessions, these rebuilds accounted for about 22% of the total bill , money spent re-encoding content that was already cached moments earlier. claude-thermos launches Claude Code behind a small local reverse proxy it points ANTHROPIC BASE URL at a loopback port; all traffic still goes to the real Anthropic API . Observe. The proxy watches /v1/messages traffic and groups it into sessions and lineages , a lineage being one cache prefix, keyed by model + tool set + system text. The first tool-bearing lineage is the main agent; the rest are subagents. Detect the danger window. When the main lineage goes idle and a subagent is actively running, the main prefix is at risk of expiring. Warm. On an interval under the 5-minute TTL, it replays the main agent's last real request as a warm request : identical cacheable prefix, but max tokens: 1 and no streaming. The single token is thrown away; the point is the prefill, which reads and refreshes the full cached prefix. Warm requests go directly to the API, never through the proxy, so they can't disturb real traffic. Result. When the subagent finishes, the main agent's cache is still warm. It pays a cheap read instead of a full rewrite. Each warm costs a cache read 0.1x ; each rewrite it prevents would have cost a write 1.25x on a much larger prefix, so the trade is heavily in your favor. Every session writes to: ~/.claude-thermos/logs/<session id / ├── events.jsonl append-only structured event stream └── summary.json rollup totals, written when the session ends events.jsonl records each request/response's token usage plus every warming decision warm fired , warm result , cap reached , resume detected , and so on . summary.json is the rollup you'll usually read: | Field | Meaning | |---|---| warms fired | Warm requests sent | cache read total | Tokens read back by those warms | episodes | Idle-with-subagent episodes that ended in a successful resume a rewrite actually avoided | rewrite avoided tokens | Tokens that would have been re-written, summed across episodes | warm cost | What warming cost you: 0.1 × cache read total | rewrite avoided cost | What it saved: 1.25 × rewrite avoided tokens | net savings | rewrite avoided cost − warm cost | All three cost figures are in base-input-token units token counts already weighted by their cache multiplier . To turn net savings into dollars, multiply it by your model's price per input token : dollars saved ≈ net savings × input token price For example, at an input price of $3 / 1M tokens, a net savings of 1 200 000 is about 1 200 000 × $3 / 1 000 000 = $3.60 saved that session.