apunkt/MetaGPT
1
0
Fork 0
mirror of https://github.com/FoundationAgents/MetaGPT.git synced 2026-07-20 16:51:07 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

修改role_zore重复指令问题

Browse source
This commit is contained in:
黄伟韬 2024-07-18 14:43:59 +08:00
parent 8922547342
commit 56efc6d155
2 changed files with 20 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

36
metagpt/roles/di/role_zero.py
View file

@ -10,9 +10,7 @@ from pydantic import model_validator
from metagpt.actions import Action, UserRequirement
from metagpt.actions.di.run_command import RunCommand
from metagpt.exp_pool import exp_cache
from metagpt.exp_pool.context_builders import RoleZeroContextBuilder
from metagpt.exp_pool.serializers import RoleZeroSerializer
from metagpt.const import DEFAULT_WORKSPACE_ROOT
from metagpt.logs import logger
from metagpt.prompts.di.role_zero import (
CMD_PROMPT,
@ -31,7 +29,6 @@ from metagpt.tools.tool_registry import register_tool
from metagpt.utils.common import CodeParser, any_to_str
from metagpt.utils.repair_llm_raw_output import RepairType, repair_llm_raw_output
from metagpt.utils.report import ThoughtReporter
from metagpt.const import DEFAULT_WORKSPACE_ROOT
@register_tool(include_functions=["ask_human", "reply_to_human"])
@ -164,22 +161,24 @@ class RoleZero(Role):
async with ThoughtReporter(enable_llm_stream=True) as reporter:
await reporter.async_report({"type": "react"})
self.command_rsp = await self.llm.aask(context, system_msgs=self.system_msg)
try :
# check and repair
try:
commands = CodeParser.parse_code(block=None, lang="json", text=self.command_rsp)
commands = json.loads(repair_llm_raw_output(output=commands, req_keys=[None], repair_type=RepairType.JSON))
except :
logger.warning('Trying to repair json string with repair tool...')
except:
logger.warning("Trying to repair json string with repair tool...")
commands = CodeParser.parse_code(block=None, lang="json", text=self.command_rsp).strip()
if commands.endswith(']') and not commands.startswith('['):
commands = '['+ commands
self.command_rsp = f"```json\n{self.commands}\n```"
if commands.endswith("]") and not commands.startswith("["):
commands = "[" + commands
self.command_rsp = f"```json\n{self.command_rsp}\n```"
print(self.command_rsp)
self.rc.memory.add(AIMessage(content=self.command_rsp))
self.rc.memory.add(AIMessage(content=self.command_rsp))
return True
async def add_editor_root_directory(self,memory) -> List[Message]:
memory.append(UserMessage(cause_by="editory", content=f'Root directory is {DEFAULT_WORKSPACE_ROOT}'))
async def add_editor_root_directory(self, memory) -> List[Message]:
memory.append(UserMessage(cause_by="editory", content=f"Root directory is {DEFAULT_WORKSPACE_ROOT}"))
return memory
async def parse_browser_actions(self, memory: List[Message]) -> List[Message]:
if not self.browser.is_empty_page:
pattern = re.compile(r"Command Browser\.(\w+) executed")
@ -257,6 +256,7 @@ class RoleZero(Role):
)
return rsp_msg
async def _parse_commands(self) -> Tuple[List[Dict], bool]:
"""Retrieves commands from the Large Language Model (LLM).
@ -294,7 +294,7 @@ class RoleZero(Role):
# handle special command first
if self._is_special_command(cmd):
special_command_output = await self._run_special_command(cmd)
outputs.append(output+':'+special_command_output)
outputs.append(output + ":" + special_command_output)
continue
# run command as specified by tool_execute_map
if cmd["command_name"] in self.tool_execution_map:
@ -318,10 +318,10 @@ class RoleZero(Role):
outputs = "\n\n".join(outputs)
return outputs
def _is_special_command(self,cmd) -> bool:
def _is_special_command(self, cmd) -> bool:
return cmd["command_name"] in self.special_tool_commands
async def _run_special_command(self, cmd) -> str:
"""command requiring special check or parsing"""
command_output = ""