ceos/unsloth
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Studio: keep local RAG answers grounded; cap tools and steer off web when RAG is on

Browse source
This commit is contained in:
Roland Tannous 2026-05-29 15:42:32 +04:00
commit 1a31aa5b27
1 changed files with 17 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

20
studio/frontend/src/features/chat/api/chat-adapter.ts
View file

@ -1743,7 +1743,13 @@ export function createOpenAIStreamAdapter(): ChatModelAdapter {
'LITERAL id attribute — e.g. `<chunk id="7">` is cited as ' +
"`[7]`. IDs are unique across the whole turn; never renumber, " +
"never reuse a different number. Do not make additional tool " +
"calls beyond the planned 3.",
"calls beyond the planned 3.\n" +
"4. Answer primarily from the knowledge base. Only use web " +
"search or web fetch if the returned chunks genuinely do not " +
"contain the answer — do not use them to double-check or expand " +
"an answer the documents already support.\n" +
"5. Once you have written your grounded answer, STOP. Do not " +
"make further tool calls and do not restate or re-answer.",
);
}
if (systemPromptParts.length > 0) {
@ -2511,8 +2517,16 @@ export function createOpenAIStreamAdapter(): ChatModelAdapter {
mcp_enabled: mcpEnabledForChat,
auto_heal_tool_calls:
useChatRuntimeStore.getState().autoHealToolCalls,
max_tool_calls_per_message:
useChatRuntimeStore.getState().maxToolCallsPerMessage,
// With RAG active the model only needs up to 3 retrieval
// calls; cap low so it can't spiral into web search / fetch
// after already answering from the documents. Off-RAG turns
// keep the user's full budget.
max_tool_calls_per_message: ragToolPathTaken
? Math.min(
useChatRuntimeStore.getState().maxToolCallsPerMessage,
6,
)
: useChatRuntimeStore.getState().maxToolCallsPerMessage,
tool_call_timeout: (() => {
const mins = useChatRuntimeStore.getState().toolCallTimeout;
return mins >= 9999 ? 9999 : mins * 60;