Skip to content

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.

Commits on main

Merge branch 'chopratejas:main' into main
JD Davis committed 6 months ago
e6a9604
28a0e19
ci: skip live evals without credentials
JerrettDavis committed 6 months ago
2fd4826
fix: populate Codex WS dashboard performance metrics
Gili Tzabari committed 6 months ago
4279a7f
fix(proxy): bound Codex Responses compression work
Gili Tzabari committed 6 months ago
160989c
Fix Windows ORT builds and Docker signing retries
Tejas Chopra committed 6 months ago
c836877
Format OpenAI Responses tests
Tejas Chopra committed 6 months ago
ead07cc
Reduce Codex compression proxy logging
Tejas Chopra committed 6 months ago
4fbfc02
Compress Codex Responses payloads
Tejas Chopra committed 6 months ago
478a75f
fix: log Codex ws cancellations safely
Tejas Chopra committed 6 months ago
62a1f23
fix: register serena mcp during wrap
Tejas Chopra committed 6 months ago
a7160f7
Fix learn CLI worker expectation in CI
Tejas Chopra committed 6 months ago
22e750c
fix: stabilize learn write-failure worker test
Tejas Chopra committed 6 months ago
155c069
fix: make proxy upgrades version-aware
Tejas Chopra committed 6 months ago
ea1f608
fix: harden learn path handling across platforms
Tejas Chopra committed 6 months ago
5ceca13
17ffae0
Merge branch 'main' into fix/codex-wrap-proxy-mcp-sync
Tejas Chopra committed 6 months ago
c7af830
fix: align docs and prune dead compatibility surfaces
Tejas Chopra committed 6 months ago
359e8c9
eaf5980
fix: sync Codex MCP proxy config during wrap
Tejas Chopra committed 6 months ago
ac1d11c