mtfyun/ragflow
1
0 Fork 0
mirror of https://github.com/infiniflow/ragflow.git synced 2026-11-11 22:01:50 +08:00
Code Issues Packages Projects Releases Wiki Activity
9,247 Commits 3 Branches 194 Tags
0a00a732905dfd6c04c2873ddafc1904a73132a2
Commit Graph
180 Commits
Author SHA1 Message Date
Yingfeng 4e54fe1338 Port Agentic RAG to Golang (#19563) 2026-09-14 09:38:51 +08:00
天海蒼灆 ebbdd5028a feat(agent): support document_ids in retrieval node and tool (#16938) 2026-09-11 19:15:23 +08:00
xugangqiang a36615c6fa Migrate cache backend from Valkey 8 to apache/kvrocks:2.16.0 (#19374) 2026-09-11 17:39:11 +08:00
qinling0210 ce34ec42f5 Port agentic RAG PRs to GO (#19527) 
### Summary

Port
https://github.com/infiniflow/ragflow/pull/19424/
https://github.com/infiniflow/ragflow/pull/19459/

Address comments from https://github.com/infiniflow/ragflow/pull/19500
 2026-09-11 15:58:42 +08:00
Sbaaoui Idriss 681f418424 fix(invoke): tolerate invalid headers across runtimes (#19481) 2026-09-11 10:21:25 +08:00
mkaaad c51f44bab3 fix(agent/tool): translate MCP tool inputSchema.properties in Info() (#19244) 
Fix MCPToolAdapter.Info() — now reads parameter names from inputSchema["properties"] instead of top-level schema keys. Model gets correct param names/types/required flags.
 2026-09-10 20:04:15 +08:00
qinling0210 ef800ba5c2 Port agentic RAG to GO (#19500) 
### Summary

Port agentic RAG to GO
 2026-09-10 20:02:48 +08:00
Sbaaoui Idriss 11f4d11920 fix(agent): preserve Starter values beside query inputs (#19420) 2026-09-10 10:16:32 +08:00
Wang Qi 2284c8e1ab Refactor: Follow #19404 to add RERANK_TOKEN_LIMIT_MODE=truncate|passthrough|raise_error (#19421) 2026-09-09 16:32:08 +08:00
Lynn cd054ed786 Fix: check sql component required fields (#19418) 2026-09-09 13:37:06 +08:00
euvre 38c40e64a9 fix(agent): treat missing List Operations input as empty list (#19275) 2026-09-09 11:07:43 +08:00
Sbaaoui Idriss d796dcfb6f fix(agent): preserve sandbox artifacts through streaming (#19355) 2026-09-09 09:57:34 +08:00
WOLIKIMCHENG 440c2c646b fix(agent): flatten scalar-list values in DataOperations combine (#19111) 2026-09-08 15:39:15 +08:00
buua436 c0470b8594 fix: support agent input mapping and converged edges (#19317) 2026-09-08 11:34:27 +08:00
euvre a41d0d3a94 fix(agent): accept user_id node param on the Retrieval agent tool (#19090) 2026-09-03 19:37:10 +08:00
jay77721andCommandCodeBot 45c3a70c2f refactor(go): merge messagefit into internal/tokenizer (#19225) 
Co-authored-by: CommandCodeBot <noreply@commandcode.ai>
 2026-09-03 19:33:29 +08:00
Sbaaoui Idriss ec8f5e0421 fix(agent): reject empty message rows when saving canvas (#19149) 2026-09-02 15:42:21 +08:00
euvre bba1b9d91b fix(agent): restore message download button when output file type is selected (#18955) 2026-09-02 14:42:03 +08:00
Sbaaoui Idriss 01cf9c428a fix error where returns a no dataset found error in a 500 wrapper (#19059) 2026-09-01 20:14:48 +08:00
euvre b54e0e0588 fix(agent): reject only truly-empty delimiter entries on canvas save (#19048) 2026-09-01 17:14:15 +08:00
Yingfeng 862ee95df2 Refine agentic search (#18798) 2026-08-28 23:56:14 +08:00
Sbaaoui Idriss 52fe959e7f fix(agent): make user added emppty entries fail to save (#18979) 2026-08-28 19:25:27 +08:00
Sbaaoui Idriss 9efdfb2c24 fix(agent): reject fractional integer DSL parameters (#18964) 2026-08-28 15:20:19 +08:00
buua436 7d658ccd17 fix: map agent rounds to react steps (#18868) 2026-08-27 19:15:31 +08:00
SYED ALI ABBAS RAHIL 82e4147b83 fix(api): honor dialog system prompt in reasoning-enabled chat (#18842) 
### Problem

When a chat assistant has a configured system prompt and the reasoning
level is set to anything other than `NONE`, the request goes through
`rag_agent()` / the agentic RAG harness. That path generated the final
answer with a fixed `FINAL_ANSWER_SYSTEM` prompt and never injected the
user-configured system prompt, so the assistant ignored
role/tone/language instructions (reported in #18833).
 2026-08-27 14:53:06 +08:00
Wang Qi ba625dc52b Refactor: refine error messages - use %W (#18834) 2026-08-27 11:21:17 +08:00
Jin Hai 6c2bca893c Go: fix context, part3 (#18821) 
Signed-off-by: Jin Hai <haijin.chn@gmail.com>
 2026-08-26 22:33:10 +08:00
陈志谦andjindou fdf937177d fix(security): authenticate sandbox executor API and isolate container network (#18703) 
Fix sandbox security gaps: (1) add token auth + rate limiter to POST /run, bind port to localhost; (2) default containers to --network none. Backward compatible. Tests pass, docs updated.

---------

Co-authored-by: jindou <jindou@local>
 2026-08-26 19:26:28 +08:00
euvre 9a03f708ec fix(agent): panic and raw error when auto play runs without a default TTS model (#18734) 2026-08-25 19:34:26 +08:00
Wang Qi 158d672c5e Porting #18737 to go (#18745) 
Porting #18737 to go
Test: passed.
 2026-08-25 18:58:34 +08:00
Guofang.Tang 80f74216d8 fix(agent): redact internal terminal errors (#18743) 2026-08-25 17:46:51 +08:00
Lem0nTea2002 bb488db9d4 fix(agent): bound crawler response bodies to 4 MiB (#18677) 2026-08-25 13:59:13 +08:00
Wang Qi 4c6f745c1e Refactor: rename prefetch_size to rerank_candidates (#18718) 2026-08-25 11:35:53 +08:00
Guofang.Tang da68cad1a0 fix(agent): surface cancellation terminal state (#18579) 
Follow up on #18316 by preserving cancellation as a terminal state in the Go Agent OpenAI-compatible response path. Previously, a cancelled run closed the event channel without an event, so the adapter could return an empty successful completion or finish a stream with stop.
 2026-08-25 10:59:29 +08:00
Wang Qi cc7d99e288 Fix: porting #18300 to go (#18690) 2026-08-24 19:00:27 +08:00
buua436 f37b6fc439 fix: account for checkpointed ingestion progress (#18667) 
Preserve completed component progress when an ingestion task resumes from a checkpoint. Checkpointed operators are skipped and do not emit lifecycle events again, which previously caused the document progress to
remain below 100%.
 2026-08-24 14:28:05 +08:00
Jin Hai 23a071157d Revert "port smart-reasoning agentic RAG to eino ADK" (#18670) 
Reverts infiniflow/ragflow#18654
 2026-08-24 12:03:11 +08:00
Zhichang Yu 2db8eb6c91 port smart-reasoning agentic RAG to eino ADK (#18654) 
Ports the smart-reasoning (agentic RAG) conversation mode to the eino ADK, with Go retrieval tools (grep_chunks, search_chunks, list_chunks), deep-read XML output, and frontend agent-mode wiring.
 2026-08-23 20:54:40 +08:00
Loong c15241c3a9 fix(agent): reset execution path for new sessions (#18532) 2026-08-20 22:42:55 +08:00
buua436 a3df588951 feat: improve incremental Wiki compilation (#18557) 2026-08-20 16:45:34 +08:00
Jack 84bed4dec5 refactor(chunker): replace QAChunker/PresentationChunker with PairChunker/PageChunker, drop TagChunker (#18523) 2026-08-20 11:53:00 +08:00
D819ClaudeCursorClaude
 		c466ff3b1a fix(llm): disable DeepSeek V4 thinking by default via extra_body (#18241) 
## Summary
- Disable DeepSeek V4 thinking (chain-of-thought) by default.
- litellm 1.82.x drops `thinking: disabled`; carry the toggle through
`extra_body.thinking.type` and strip `reasoning_effort` to avoid the
400.
- Use local timezone for agent `sys.date` instead of UTC.

Reference: https://api-docs.deepseek.com/guides/thinking_mode

---------

Co-authored-by: Claude <claude@example.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
Co-authored-by: Claude <noreply@anthropic.com>
 2026-08-19 13:47:36 +08:00
Jin Hai 68f8260821 Go: fix context, part3 (#18390) 
Signed-off-by: Jin Hai <haijin.chn@gmail.com>
 2026-08-18 00:07:37 +08:00
questfever 52519e2fd7 refactor: replace Split in loops with more efficient SplitSeq (#18248) 2026-08-17 19:04:04 +08:00
euvre 3d50f8bcfc fix(go-agent): forward sys.files uploads to vision LLMs (#18246) 2026-08-14 18:26:09 +08:00
Jin Hai 2c775d0549 Go: fix context (#18291) 
Signed-off-by: Jin Hai <haijin.chn@gmail.com>
 2026-08-14 17:50:13 +08:00
Zhichang Yu 620f807cb5 Port agentic search high/ultra research loop to Go (#18242) 
Implement two-level research loop, sufficiency ladder, AutoRater, grounded review, pipeline, inspector tools, and graph exploration for high/ultra agentic search modes.
 2026-08-14 10:16:46 +08:00
Haruko386 a6f2eac496 fix: unable to pass Temperature and MaxTokens in agent (#18223) 
### Summary

As title
 2026-08-13 18:31:08 +08:00
wenqian c589a5d050 feat: add UCloud agent sandbox provider (#18088) 2026-08-13 15:33:03 +08:00
jay77721 0a4cc65db0 refactor(common): unify </think>-strip into shared StripThinkTrailing helper (#18179) 
Collapse the nine identical **"cut through the last `</think>`"**
implementations — mirroring Python's `re.sub(r"^.*</think>", "", s,
re.DOTALL)` — into one shared helper `common.StripThinkTrailing`,
preventing future behavior drift between copies.
 2026-08-13 13:56:56 +08:00