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(install): stabilize Windows task lifecycle and eliminate console window popup (#4627)
6b903cb Vishal Tyagi committed 15 days ago
fix(proxy): say a forwarded streaming error came from the provider (#4629)
956b885 gglucass committed 15 days ago
docs(security): sync security model with egress pollers, counts and beacon payload (#4539)
ee975d4 HetCreep committed 15 days ago
fix(proxy): keep Gemini memory context on entries with non-text parts (#4506)
35a2b76 Gaurav Dubey committed 15 days ago
fix(offline): guard the learn and memory-eval LLM calls under HEADROOM_OFFLINE (#4344)
dd361bb Abel Yagubyan committed 15 days ago
fix(proxy): let model discovery honor the custom base URL (#4330)
911c75f Vaibhav Hajare committed 15 days ago
fix(tabular): keep tables whose columns share a name (#4329)
ddbbb07 Vaibhav Hajare committed 15 days ago
fix(kompress): keep words past the tokenizer's truncation point (#4328)
bb13384 Vaibhav Hajare committed 15 days ago
fix(diff): renumber hunk headers after trimming context (#4319)
9355ec7 Jay Hemnani committed 15 days ago
deps: Bump tsup from 8.5.1 to 9.0.0 in /plugins/openclaw (#4384)
9d87151 dependabot[bot] committed 15 days ago
fix(memory): preserve parent scopes when clearing memories (#4305)
a8375a1 takiAA committed 15 days ago
fix(proxy): strip only one leading space from SSE field values (#4306)
9226eaf kevin committed 15 days ago
fix(cache): reject nonpositive semantic cache capacities (#4301)
911bb74 Romin Parekh committed 15 days ago
fix(transforms): stop the dense-line elider cutting Chinese, Japanese and Thai prose (#4295)
7c4af75 Rudy Mizrahi Celekli committed 15 days ago
fix(transforms): recognise code fences whose info string is not a bare word (#4291)
a8df6f9 Rudy Mizrahi Celekli committed 15 days ago
fix(config): use the longest matching model_context_limits prefix (#4288)
603d873 Rudy Mizrahi Celekli committed 15 days ago
fix(proxy): allow loopback-published container hosts to access /debug/memory (#4259)
e546e71 lorenzozane committed 15 days ago
fix(memory): keep learned evidence until persistence succeeds (#4229)
662a7f6 Mansurshakh Japarov committed 15 days ago
fix(proxy): answer /api/hello probes behind the token gate, log rejection context (#4150)
673831a Serge ARADJ committed 15 days ago
fix(tokenizers): resolve tiktoken encoding for gateway-prefixed model ids (#4637)
e9331b6 Alvi Rahman committed 15 days ago
fix(proxy): bypass compression for large admitted histories (#4142)
50bf775 Mansurshakh Japarov committed 15 days ago
fix(openclaw): read the retrieve hash from params, not the tool-call id (#4641)
362e1fc Vivek Singh committed 15 days ago
feat(ccr): answer headroom_retrieve on streaming chat completions (#4631)
31dcaf4 SK Arin committed 15 days ago
fix(proxy): skip memory-context search on OpenAI paths in cache mode (#4634)
3654037 Luis Araya committed 15 days ago
fix(proxy): skip memory-context search in cache mode on OpenAI routes (#4632)
365f6fe AKSHITH BHARADWAJ D L T committed 15 days ago
test(tokenizer): hold the env lock in the invalid-repo Hub error test (#4151)
1bbc0ea Kenneth Wong committed 15 days ago
fix(kompress): read the pinned snapshot from Python's one HF cache in the native proxy (#4438)
eef44f6 Kenneth Wong committed 15 days ago
fix: [BUG] Volatile SQLite metrics stores lose their schema before the first save (#4633)
d8ff0fe Nandana Dileep committed 15 days ago