Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Python agentaianthropicclaude-codecompressioncontext-engineeringcontext-windowcursorfastapilangchainllmmcpopenaiprompt-engineeringproxypythonragtoken-optimizationtokenstypescript
fix: skip token inflation guard in cache mode
Tejas Chopra1006bc3522b0
diff --git a/headroom/proxy/handlers/anthropic.py b/headroom/proxy/handlers/anthropic.pyindex 365a866bc..a41210c2c 100644--- a/headroom/proxy/handlers/anthropic.py+++ b/headroom/proxy/handlers/anthropic.py@@ -681,8 +681,9 @@ class AnthropicHandlerMixin:# Flag compression failure for observability_compression_failed = True- # Guard: if "optimization" inflated tokens, revert to originals- if optimized_tokens > original_tokens:+ # Guard: if "optimization" inflated tokens, revert to originals.+ # Skip in cache mode where prefix-stability may legitimately shift counts.+ if optimized_tokens > original_tokens and not is_cache_mode(self.config.mode):logger.warning(f"[{request_id}] Optimization inflated tokens "f"({original_tokens} -> {optimized_tokens}), reverting to original messages"@@ -691,7 +692,7 @@ class AnthropicHandlerMixin:optimized_tokens = original_tokenstransforms_applied = []- tokens_saved = original_tokens - optimized_tokens+ tokens_saved = max(0, original_tokens - optimized_tokens)optimization_latency = (time.time() - start_time) * 1000# Hook: post_compress — let hooks observe compression results