Shy2593666979/AgentChat

AgentChat is an LLM-based agent communication platform with built-in default Agents and support for user-defined Agents. Through multi-turn conversations and task collaboration, Agents can understand and help complete complex tasks. The project integrates LangChain, Function Call, the MCP protocol, RAG, Memory, HITL, Skill, Milvus, and ElasticSearch for efficient knowledge retrieval and tool calling, with a high-performance backend built using FastAPI.

921Stars on the repository
6Mods indexed here, across every type
yesterdayLast push, which is what freshness is scored on
MITLicence, which decides whether bodies are shown

code-review

01

Shy2593666979/AgentChat

Skill Claude CodeCodex

A code-review procedure for examining supplied code or a change diff for evidence-based bugs, compatibility problems, and data risks.

not rated 921 +7 yesterday A 41 tokens original MIT

data-analysis

02

Shy2593666979/AgentChat

Skill Claude CodeCodex

A method for reviewing tables and comparing measurements while checking data quality. It examines fields, units, time ranges, missing and duplicate records, unusual values, and the definitions behind each metric.

not rated 921 +7 yesterday A 34 tokens original MIT

email-writing

03

Shy2593666979/AgentChat

Skill Claude CodeCodex

An email-drafting method that turns supplied facts, recipients, purpose, language, and tone into a reviewable message. It can also revise an existing draft while preserving its meaning.

not rated 921 +7 yesterday A 37 tokens original MIT

Shy2593666979/AgentChat

Skill Claude CodeCodex

A knowledge-organising method turns notes, conversations, or documents into knowledge cards, frequently asked questions, how-to guides, or decision records. It keeps sources, unresolved points, conditions, and exceptions visible.

not rated 921 +7 yesterday A 41 tokens original MIT

product-document

05

Shy2593666979/AgentChat

Skill Claude CodeCodex

A method for turning requirements into product documents, feature explanations, or user guides. It describes user scenarios, scope, normal and error flows, permissions, dependencies, and testable acceptance conditions.

not rated 921 +7 yesterday A 33 tokens original MIT

research-report

06

Shy2593666979/AgentChat

Skill Claude CodeCodex

A method for writing a research report, topic review, or comparison from supplied material. It separates evidence, source authors' views, and the writer's inferences while keeping sources identifiable.

not rated 921 +7 yesterday A 35 tokens original MIT

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: