Skip to content

清理 web_search 返回结果中的 <think> 标签 #37

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
drunkduckdrown wants to merge 3 commits into
base: grok-with-tavily
Choose a base branch
from
Open

清理 web_search 返回结果中的 <think> 标签 #37

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
0bbccff
去除Grok返回的<think>标签
drunkduckdrown Mar 19, 2026
e06808f
弱化search_prompt，避免触发Grok越狱抵抗
drunkduckdrown Mar 19, 2026
a49d6a1
Revert accidental changes from wrong branch
drunkduckdrown Mar 19, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 3 additions & 2 deletions src/grok_search/server.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,13 +15,13 @@
from grok_search.providers.grok import GrokSearchProvider
from grok_search.logger import log_info
from grok_search.config import config
from grok_search.sources import SourcesCache, merge_sources, new_session_id, split_answer_and_sources
from grok_search.sources import SourcesCache, merge_sources, new_session_id, sanitize_answer_text, split_answer_and_sources
from grok_search.planning import engine as planning_engine, _split_csv
except ImportError:
from .providers.grok import GrokSearchProvider
from .logger import log_info
from .config import config
from .sources import SourcesCache, merge_sources, new_session_id, split_answer_and_sources
from .sources import SourcesCache, merge_sources, new_session_id, sanitize_answer_text, split_answer_and_sources
from .planning import engine as planning_engine, _split_csv

import asyncio
Expand Down Expand Up @@ -203,6 +203,7 @@ async def _safe_firecrawl() -> list[dict] | None:
firecrawl_results = gathered[idx]

answer, grok_sources = split_answer_and_sources(grok_result)
answer = sanitize_answer_text(answer)
extra = _extra_results_to_sources(tavily_results, firecrawl_results)
all_sources = merge_sources(grok_sources, extra)

Expand Down
14 changes: 14 additions & 0 deletions src/grok_search/sources.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,10 @@
_SOURCES_FUNCTION_PATTERN = re.compile(
r"(?im)(^|\n)\s*(sources|source|citations|citation|references|reference|citation_card|source_cards|source_card)\s*\("
)
_THINK_TAG_PATTERN = re.compile(
r"<think\b[^>]*>.*?</think>",
re.IGNORECASE | re.DOTALL,
)


def new_session_id() -> str:
Expand Down Expand Up @@ -67,6 +71,16 @@ def merge_sources(*source_lists: list[dict]) -> list[dict]:
return merged


def sanitize_answer_text(text: str) -> str:
"""Remove model reasoning tags from answer text while preserving content."""
raw = (text or "").strip()
if not raw:
return ""
cleaned = _THINK_TAG_PATTERN.sub("", raw)
cleaned = re.sub(r"\n{3,}", "\n\n", cleaned).strip()
return cleaned


def split_answer_and_sources(text: str) -> tuple[str, list[dict]]:
raw = (text or "").strip()
if not raw:
Expand Down