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
Commits on main
fix(proxy): allow HEAD method on catch-all passthrough route (#2035)
c41cf44 GUOHAO LIU committed 4 months ago
fix(proxy): handle ClientDisconnect in passthrough body reads + log sanitization (#2067)
605e269 GUOHAO LIU committed 4 months ago
fix(proxy/gemini): thread savings-profile kwargs into apply() (#1994)
38306a3 Abhay Singh committed 4 months ago
feat(transforms): first-class C# support in CodeAwareCompressor (#1926)
ec55ddc Yevhen Koval committed 4 months ago
refactor(proxy): extract memory golden replay policy (#2007)
8c68f48 JD Davis committed 4 months ago
fix(ccr): don't crash parse_tool_call on non-object tool arguments (#2071)
984a2c7 Abhay Singh committed 4 months ago
feat(observability): add gen_ai.request.model to the compression span (#1667)
7f7af66 Krishna Chaitanya committed 4 months ago
fix(cache): partial cached-prefix replay + idle-aware net-cost; don't… (#1933)
b0440f9 Tejas Chopra committed 4 months ago
fix(mcp/opencode): don't clobber an unparseable opencode.json on register (#1661)
d079614 Abhay Singh committed 4 months ago
deps: update tree-sitter requirement from <0.26,>=0.25.2 to >=0.25.2,<0.27 (#1681)
ce3c959 dependabot[bot] committed 4 months ago
fix(update): prevent _core.pyd corruption on Windows when proxy is running (#1581)
0750bbf JoaoMarcos44 committed 4 months ago