Convex Agents Messages
About
This skill provides message management for agent conversations, allowing developers to send, save, and retrieve messages. Use it for displaying conversation history, handling user input, and implementing optimistic UI updates. It also supports UIMessages for rich rendering and filtering tool messages from the displayed history.
Documentation
Purpose
Messages represent individual exchanges in a conversation thread. This skill covers saving, retrieving, displaying, and managing messages.
When to Use This Skill
- Displaying conversation history to users
- Saving user messages before generating responses
- Implementing optimistic message updates in UI
- Working with UIMessages for rich formatting
- Filtering tool messages from displayed history
Retrieve Messages
import { listUIMessages } from "@convex-dev/agent";
export const listThreadMessages = query({
args: { threadId: v.string(), paginationOpts: paginationOptsValidator },
handler: async (ctx, { threadId, paginationOpts }) => {
return await listUIMessages(ctx, components.agent, {
threadId,
paginationOpts,
});
},
});
Display in React
import { useUIMessages } from "@convex-dev/agent/react";
function ChatComponent({ threadId }: { threadId: string }) {
const { results } = useUIMessages(
api.messages.listThreadMessages,
{ threadId },
{ initialNumItems: 20 }
);
return (
<div>
{results?.map((message) => (
<div key={message.key}>{message.text}</div>
))}
</div>
);
}
Save a Message Manually
import { saveMessage } from "@convex-dev/agent";
export const saveUserMessage = mutation({
args: { threadId: v.string(), prompt: v.string() },
handler: async (ctx, { threadId, prompt }) => {
const { messageId } = await saveMessage(ctx, components.agent, {
threadId,
prompt,
skipEmbeddings: true,
});
return { messageId };
},
});
Optimistic Updates
Show messages immediately before they're saved:
const sendMessage = useMutation(api.chat.generateResponse).withOptimisticUpdate(
optimisticallySendMessage(api.messages.listThreadMessages)
);
Delete Messages
await myAgent.deleteMessage(ctx, { messageId });
await myAgent.deleteMessages(ctx, { messageIds });
await myAgent.deleteMessageRange(ctx, { threadId, startOrder, endOrder });
UIMessage Type
Extended message with agent-specific fields:
interface UIMessage {
key: string;
role: "user" | "assistant" | "system";
text: string;
status: "saved" | "streaming" | "finished" | "aborted";
agentName?: string;
_creationTime?: Date;
parts?: MessagePart[];
}
Key Principles
- UIMessages vs MessageDocs: Use UIMessages for UI, MessageDocs for backend
- Message ordering: Maintained automatically by order and stepOrder
- Optimistic updates: Better UX showing messages before save
- Skip embeddings in mutations: Use when saving in mutations
Next Steps
- See streaming for real-time message updates
- See files for attaching media to messages
- See threads for managing message containers
Quick Install
/plugin add https://github.com/Sstobo/convex-skills/tree/main/convex-agents-messagesCopy and paste this command in Claude Code to install this skill
GitHub 仓库
Related Skills
langchain
MetaLangChain is a framework for building LLM applications using agents, chains, and RAG pipelines. It supports multiple LLM providers, offers 500+ integrations, and includes features like tool calling and memory management. Use it for rapid prototyping and deploying production systems like chatbots, autonomous agents, and question-answering services.
Algorithmic Art Generation
MetaThis skill helps developers create algorithmic art using p5.js, focusing on generative art, computational aesthetics, and interactive visualizations. It automatically activates for topics like "generative art" or "p5.js visualization" and guides you through creating unique algorithms with features like seeded randomness, flow fields, and particle systems. Use it when you need to build reproducible, code-driven artistic patterns.
webapp-testing
TestingThis Claude Skill provides a Playwright-based toolkit for testing local web applications through Python scripts. It enables frontend verification, UI debugging, screenshot capture, and log viewing while managing server lifecycles. Use it for browser automation tasks but run scripts directly rather than reading their source code to avoid context pollution.
requesting-code-review
DesignThis skill dispatches a code-reviewer subagent to analyze code changes against requirements before proceeding. It should be used after completing tasks, implementing major features, or before merging to main. The review helps catch issues early by comparing the current implementation with the original plan.
