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
Add hnswlib to dev dependencies for CI testing
chopratejasdcae408dcd4b
diff --git a/pyproject.toml b/pyproject.tomlindex f2b6752a2..03e218326 100644--- a/pyproject.toml+++ b/pyproject.toml@@ -90,6 +90,10 @@ evals = ["anthropic>=0.18.0", # Anthropic API for evals"openai>=1.0.0", # OpenAI API for evals]+# Memory system extras (hierarchical memory with vector search)+memory = [+ "hnswlib>=0.8.0", # HNSW vector index for semantic search+]# Development dependenciesdev = ["pytest>=7.0.0",@@ -102,10 +106,11 @@ dev = ["anthropic>=0.18.0","ollama>=0.4.0", # For Ollama integration tests (local LLM, no API key needed)"langchain-ollama>=0.2.0", # For LangChain+Ollama integration tests+ "hnswlib>=0.8.0", # For memory system tests]# All optional dependenciesall = [- "headroom-ai[relevance,proxy,reports,llmlingua,code,evals]",+ "headroom-ai[relevance,proxy,reports,llmlingua,code,evals,memory]",][project.scripts]