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
feat(proxy): apply output shaper to OpenAI-compatible endpoints (#1725)
e65b9b3 Dylan Russell, MD committed 4 months ago
fix: add Vercel deploy config and workflow for docs site (#1739)
b0fa84e Gaurav Yadav committed 4 months ago
docs: add Claude Code status-line indicator to Community (#1790)
b4d4f64 Ship-Wright committed 4 months ago
feat(evals): register multilingual multi-wiki-qa (zh/ja/ko) dataset (#1530)
f891506 Zhenjia ZHOU committed 4 months ago
test(codex-ws): de-flake semaphore-tail test (widen noise floor 25ms->75ms) (#1736)
9446eea Tejas Chopra committed 4 months ago
fix(proxy/batch): preserve sibling tool configs on Google batch requests (#2177)
a5d7e12 Abhay Singh committed 4 months ago
fix(savings): record pre-compression original as ledger before, not forwarded count (#2176)
195ed90 Abhay Singh committed 4 months ago
fix(proxy/gemini): forward a non-JSON upstream body with its real status (#2174)
f723925 Abhay Singh committed 4 months ago
fix(install): guard non-dict health config in 'install status' (#2150)
8f867e4 Abhay Singh committed 4 months ago
fix(telemetry): only advance usage-report baseline after a 200 (#2149)
0cddac6 Abhay Singh committed 4 months ago
fix(savings): don't bill free models at the $3/M fallback in the ledger (#2147)
fb17156 Abhay Singh committed 4 months ago
fix(proxy/savings): append history point on cache-only savings too (#2194)
d125805 Ingmar Krusch committed 4 months ago
fix(init/codex): merge into hooks.json instead of overwriting it (#2173)
84f66da Abhay Singh committed 4 months ago
fix(init/codex): don't delete per-profile provider settings (#2146)
8da4384 Abhay Singh committed 4 months ago
fix(proxy): reject rate_limit_requests_per_minute=0 when limiting is enabled (#2142)
8a71947 Abhay Singh committed 4 months ago
fix(learn): don't desync verbosity pairing on empty assistant turns (#2123)
def2f9a Abhay Singh committed 4 months ago
fix(wrap/claude): bind _wrap_settings_path before the try (#2126)
faed4dc Abhay Singh committed 4 months ago
fix(wrap/codex): export the detected custom upstream base URL (#2125)
d236b27 Abhay Singh committed 4 months ago
fix(cache): normalize embeddings before the semantic similarity check (#2122)
f8eaaeb Abhay Singh committed 4 months ago
fix(litellm): forward chat_template_kwargs and other vendor top-level fields to OpenAI-compatible backends via extra_body (#2128) (#2163)
fb683e1 Rod Boev committed 4 months ago
fix(cache-aligner): hash the frozen conversation prefix so Claude Code cache invalidation is detected (#2085) (#2161)
cc072f0 Rod Boev committed 4 months ago
docs: qualify CCR auto-resolution support for Gemini (#2044)
e6df6ea Rod Boev committed 4 months ago
fix(codex): rewrite config.toml properly so Codex will route through … (#2102)
5d9bbbe nangsontay committed 4 months ago
fix(subscription): keep efficiency_pct from exceeding 100% (#2121)
5fb449e Abhay Singh committed 4 months ago
fix(mcp/claude): don't clobber an unparseable Claude config on register (#1660)
bc24e25 Abhay Singh committed 4 months ago
docs: document Claude VSCode deferred-tool rendering caveat (#2045)
81ddbd4 Rod Boev committed 4 months ago
install: couple Codex routing to persistent runtime readiness (#2043)
3a39cb9 Rod Boev committed 4 months ago