wuyang/zk-data-agent
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Remove legacy online mining tools

Browse Source
This commit is contained in:
wuyang6
2026-05-12 18:03:39 +08:00
parent 6dad2cc4b3
commit a48c86501a
4 changed files with 6 additions and 1011 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/agent_tools.py
-97
View File
@@ -51,13 +51,6 @@ from .data_agent_inputs import (
load_input_sources,
render_source_context,
)
from .data_agent_router_sessions import (
DataAgentRouterSessionError,
convert_router_candidates_to_records,
profile_router_sessions,
sample_router_candidates,
search_router_sessions,
)
from .session_env_vars import get_session_env_vars
@@ -1041,10 +1034,6 @@ def default_tool_registry() -> dict[str, AgentTool]:
'data_agent_load_input_sources': _data_agent_load_input_sources_tool,
'data_agent_extract_case_evidence': _data_agent_extract_case_evidence_tool,
'data_agent_render_source_context': _data_agent_render_source_context_tool,
'data_agent_profile_router_sessions': _data_agent_profile_router_sessions_tool,
'data_agent_search_router_sessions': _data_agent_search_router_sessions_tool,
'data_agent_sample_router_candidates': _data_agent_sample_router_candidates_tool,
'data_agent_convert_router_candidates_to_records': _data_agent_convert_router_candidates_to_records_tool,
'data_agent_show_generation_plan': _data_agent_show_generation_plan_tool,
'data_agent_update_generation_plan': _data_agent_update_generation_plan_tool,
'data_agent_confirm_generation_plan': _data_agent_confirm_generation_plan_tool,
@@ -1350,92 +1339,6 @@ def _data_agent_render_source_context_tool(arguments: dict[str, Any], context: T
return data_agent_dumps_payload(payload)
def _data_agent_profile_router_sessions_tool(arguments: dict[str, Any], context: ToolExecutionContext) -> str:
try:
payload = profile_router_sessions(
context.root,
dates=_optional_string_list(arguments, 'dates'),
paths=_optional_string_list(arguments, 'paths'),
max_files=_coerce_int(arguments, 'max_files', 8),
max_rows_per_file=_coerce_int(arguments, 'max_rows_per_file', 2000),
examples_limit=_coerce_int(arguments, 'examples_limit', 3),
)
except DataAgentRouterSessionError as exc:
raise ToolExecutionError(str(exc)) from exc
return data_agent_dumps_payload(payload)
def _data_agent_search_router_sessions_tool(arguments: dict[str, Any], context: ToolExecutionContext) -> str:
try:
payload = search_router_sessions(
context.root,
dates=_optional_string_list(arguments, 'dates'),
paths=_optional_string_list(arguments, 'paths'),
devices=_optional_string_list(arguments, 'devices'),
domains=_optional_string_list(arguments, 'domains'),
intents=_optional_string_list(arguments, 'intents'),
funcs=_optional_string_list(arguments, 'funcs'),
query_keywords=_optional_string_list(arguments, 'query_keywords'),
keyword_match_mode=_optional_string(arguments, 'keyword_match_mode') or 'any',
query_regex=_optional_string(arguments, 'query_regex'),
match_scope=_optional_string(arguments, 'match_scope') or 'any_turn',
turn_count_min=_optional_int(arguments, 'turn_count_min'),
turn_count_max=_optional_int(arguments, 'turn_count_max'),
max_files=_coerce_int(arguments, 'max_files', 20),
max_rows_per_file=_coerce_int(arguments, 'max_rows_per_file', 5000),
max_candidates=_coerce_int(arguments, 'max_candidates', 200),
)
except DataAgentRouterSessionError as exc:
raise ToolExecutionError(str(exc)) from exc
return data_agent_dumps_payload(payload)
def _data_agent_sample_router_candidates_tool(arguments: dict[str, Any], _context: ToolExecutionContext) -> str:
if 'candidates' not in arguments:
raise ToolExecutionError('candidates is required')
try:
payload = sample_router_candidates(
arguments['candidates'],
sample_size=_coerce_int(arguments, 'sample_size', 100),
strategy=_optional_string(arguments, 'strategy') or 'random',
seed=_coerce_int(arguments, 'seed', 20260430),
)
except DataAgentRouterSessionError as exc:
raise ToolExecutionError(str(exc)) from exc
return data_agent_dumps_payload(payload)
def _data_agent_convert_router_candidates_to_records_tool(
arguments: dict[str, Any],
_context: ToolExecutionContext,
) -> str:
if 'candidates' not in arguments:
raise ToolExecutionError('candidates is required')
review_decisions = arguments.get('review_decisions')
if review_decisions is not None and not isinstance(review_decisions, list):
raise ToolExecutionError('review_decisions must be an array')
include_uncertain = arguments.get('include_uncertain', False)
if not isinstance(include_uncertain, bool):
raise ToolExecutionError('include_uncertain must be a boolean')
default_complex = arguments.get('default_complex', False)
if not isinstance(default_complex, bool):
raise ToolExecutionError('default_complex must be a boolean')
try:
payload = convert_router_candidates_to_records(
arguments['candidates'],
dataset_label=_require_string(arguments, 'dataset_label'),
default_target=_optional_string(arguments, 'default_target'),
default_complex=default_complex,
review_decisions=review_decisions,
batch_id=_optional_string(arguments, 'batch_id') or 'router',
include_uncertain=include_uncertain,
)
payload['validation'] = validate_dataset_records(payload['records'])
except (DataAgentRouterSessionError, DataRecordError) as exc:
raise ToolExecutionError(str(exc)) from exc
return data_agent_dumps_payload(payload)
def _data_agent_show_generation_plan_tool(arguments: dict[str, Any], context: ToolExecutionContext) -> str:
try:
payload = get_generation_plan(