Build and run agents you can see, understand and trust.
Python agentchatbotlarge-language-modelsllmllm-agentmcpmulti-agentmulti-modalreact-agentrealtime-agentvoice-agentvoice-assistant
Commits on main
refactor(workspace): add e2b and docker workspace, and the web ui & example (#1650)
22ac303 DavdGao committed 5 months ago
fix(tool): close .env bypass gap and refine dangerous-path API for Write/Edit/Bash (#1656)
123427d qbc committed 5 months ago
fix(formatter): download remote image URLs to base64 for Moonshot API (#1653)
c9b36a1 qbc committed 5 months ago
feat(model): fix dashscope structured output and add examples (#1651)
9e532db qbc committed 5 months ago
feat(tool): integrate the tool and workspace modules with the Agent class (#1642)
cc1087d DavdGao committed 6 months ago
feat(msg): add usage field in Msg class to record the token usage (#1639)
145bee0 qbc committed 6 months ago
fix(mcp): preserve $defs in MCPTool input schema and strip titles recursively (#1595)
eefa532 qbc committed 6 months ago
fix(scripts): assign a list of TextBlock to the content instead of a string (#1629)
595d2a5 qbc committed 6 months ago
fix(model): refine _format_tools for openai response model (#1635)
cda90db qbc committed 6 months ago
feat(model): unify the thinking related argument for the chat model classes (#1622)
d082682 qbc committed 6 months ago
feat(tracing): add OpenTelemetry-based tracing as a middleware (#1633)
e572c5d qbc committed 6 months ago
feat(service): implement the FastAPI based agent service (#1568)
a899b09 DavdGao committed 6 months ago
refactor(tracing): refactor the tracing module into an agent middleware (#1579)
aa3e5ad qbc committed 6 months ago
refactor(model): refactor dashscope model to openai compatible (#1617)
ec0a6e2 qbc committed 6 months ago
feat(model): add cache_creation_input_tokens and cache_input_tokens in ChatUsage (#1602)
8a32a36 qbc committed 6 months ago
feat(scripts): add test scripts for chat models in agentscope with various input (#1604)
b6be027 qbc committed 6 months ago
factor(tool): refactor the `tool_choice` argument to support advanced prompt cache in OpenAI response API (#1524)
03e7c7c qbc committed 6 months ago
refactor(model): refactor the chat model implementation (#1564)
a063cdb DavdGao committed 6 months ago
feat(workspace): built the workspace module in agentscope 2.0 (#1586)
9f6980c DavdGao committed 6 months ago
feat(context): support to compact the tool result within the `Agent` class (#1585)
54dbead DavdGao committed 6 months ago
feat(middleware): support 2.0 middlewares in the agent class (#1565)
c7e20ce DavdGao committed 6 months ago
refactor(mcp): unify the MCP implementation into a `MCPClient` class (#1572)
dc768cb DavdGao committed 6 months ago
refactor(mcp): rename the MCP tools when registered into the toolkit (#1552)
5ca2e02 DavdGao committed 6 months ago
feat(context): support context compression in Agent class (#1544)
e7dfee0 DavdGao committed 6 months ago
feat(agent): implement permission checking logic within the agent class (#1518)
0f4243b DavdGao committed 6 months ago