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(memory): honor explicit store=false on Responses requests (#2017)
31abb69 Rod Boev committed 4 months ago
fix(proxy/memory): capture user text blocks for the retrieval query (#2064)
f542b70 Abhay Singh committed 4 months ago
fix(memory/sqlite): don't emit OFFSET without LIMIT in query (#2063)
a5bdc54 Abhay Singh committed 4 months ago
fix(proxy/gemini): preserve non-text content across the compression round-trip (#2079)
4056117 Abhay Singh committed 4 months ago
fix(cache/ccr): don't evict a live entry on a duplicate store at capacity (#2082)
1138946 Abhay Singh committed 4 months ago
fix(tokenizers): recurse into list-content tool_result blocks (#2081)
dfb1d37 Abhay Singh committed 4 months ago
fix(version): mark source-checkout builds as -dev (#2072)
1cc9979 Tejas Chopra committed 4 months ago
fix(proxy/cost): price cache savings by most-used model, not first-seen (#2023)
b4f807f Abhay Singh committed 4 months ago
fix(mcp/codex): don't clobber an unparseable/non-table config.toml (#2062)
415e03c Abhay Singh committed 4 months ago
fix(install): add orjson to [proxy] extra for LiteLLM provider backends (#2074)
4f3d5ab 石岳峰 committed 4 months ago
feat(wrap): allow project RTK instruction opt-out (#2078)
f53f720 Andrew Barnes committed 4 months ago
feat(proxy): persist per-model savings breakdown in proxy_savings.json (#2055)
12a38d3 GUOHAO LIU committed 4 months ago
fix(health): exclude kompress from aggregate readiness + adversarial PBT (#2066)
f1663ea GUOHAO LIU committed 4 months ago
fix(proxy): handle content-part outputs in Codex Responses compression (#2052)
c9a7755 GUOHAO LIU committed 4 months ago
fix(tokenizers): price CJK in the fixed-ratio estimator path (#2080)
cd3d5aa Abhay Singh committed 4 months ago
fix(proxy): only queue mid-turn messages for opt-in clients with explicit session header (#1951)
c365c7f wzy-del committed 4 months ago
feat(proxy): extend output shaping to the OpenAI Responses path (Codex HTTP + WS) (#1943)
71cbb6a alex33d committed 4 months ago
fix(proxy/vertex): route google-publisher requests to the request region (#2069)
1843346 Abhay Singh committed 4 months ago
fix(proxy/openai): respect explicit stream_options.include_usage (#2026)
19201e8 Abhay Singh committed 4 months ago
fix(proxy/anthropic): scope session id by top-level system prompt (#2070)
ec6e60e Abhay Singh committed 4 months ago
fix(subscription/copilot): preserve remaining=0 for exhausted quota (#1997)
cbb7750 Abhay Singh committed 4 months ago
feat(proxy): report new-content-relative input savings rate in /stats (#2058)
112d95b gglucass committed 4 months ago
fix(install): only validate requested targets on the manual path (#1659)
18e5680 Abhay Singh committed 4 months ago
fix(dashboard): serve per-request metadata to trusted-gateway peers (#1766)
560319c Kenneth Wong committed 4 months ago