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
Merge pull request #236 from gglucass/fix/streaming-anthropic-logger
f5e8bea Tejas Chopra committed 7 months ago
test(learn): cover flush_to_file, backend edge cases, and hydrate/bump error paths
b2536e6 Garm committed 7 months ago
test(proxy): warm up TestClient before measuring /livez backpressure latency
bde7807 Garm committed 7 months ago
Merge remote-tracking branch 'upstream/main' into fix/streaming-anthropic-logger
934345d Garm committed 7 months ago
Merge remote-tracking branch 'upstream/main' into fix/learn-preserve-prior-recommendations
aa5e5ae Garm committed 7 months ago
Merge remote-tracking branch 'upstream/main' into fix/traffic-learner-evidence-count
c571fd3 Garm committed 7 months ago
Merge pull request #235 from gglucass/fix/cli-submodule-binding
706c5e9 Tejas Chopra committed 7 months ago
feat(learn): live flush of traffic patterns to agent-native context files
d9138a3 Garm committed 7 months ago
Merge remote-tracking branch 'upstream/main' into fix/dashboard-stats-snapshot-cache
d17aea2 Kayzo committed 7 months ago
fix(tests): scope sys.modules mutation and unpin hardcoded version
1788d90 Garm committed 7 months ago