Agent tool timeline

Plugins compose models, tools, sessions, sandboxes and execution loops, with a recorded task trajectory.

A persistent agent combines memory, reusable skills and tools to carry out tasks locally or on a server.

The personal-agent project adopts the OpenClaw name across messaging and local tools.

Claude Code-style execution expands to files, documents, and office work.

A minimal coding harness emphasizes tools, context control, and session management.

Agents work across the editor, terminal and browser, presenting artifacts that document progress and verification.

An open-source coding agent connects to different models to read files, edit code and use project tools.

Claude Code SDK becomes Claude Agent SDK, expanding its focus to general-purpose agents.

A reusable harness combines planning, a filesystem, and subagents for longer tasks.

Research and action come together through a virtual computer, browser, and terminal.

An IDE organizes requirements, design and task lists before agents edit code, with event-triggered automation.

Google releases an open-source terminal agent for coding, troubleshooting, and tools.

An asynchronous agent reads repositories, edits code and runs tests in a cloud VM, then presents changes for review.

Cloud sandboxes support parallel engineering tasks and reviewable code changes.

An open-source terminal coding agent works directly with local code and tools.

Agent orchestration adds handoffs, guardrails, and tracing around model and tool calls.

A terminal agent reads code, edits files, runs tests, and uses command-line tools.

Multi-step web research analyzes sources and produces cited reports.

A local desktop and command-line agent carries out tasks through connected tools and extensions.

Models express actions as code and use tools and execution environments to complete multi-step tasks.

An open protocol standardizes connections between models, data, and tools.

Composer gains an early agent that retrieves context and uses the terminal.

An agentic editor combines project context with multi-step code changes.

Claude Dev becomes Cline, with streamed edits, mid-task feedback and broader model support.

An execution platform connects coding agents to files, terminals, and browsers.

A sandboxed agent plans engineering tasks using an editor, terminal, and browser.

Stateful graphs and cycles organize controllable agent execution.

Orchestrate workflows through conversations between models, tools, and people.

Terminal-based code editing uses a repository map to give the model context across files.

An early autonomous agent uses repeated model calls and tools, adding Selenium-based web browsing in this release.