> ## Documentation Index > Fetch the complete documentation index at: https://ade-app.dev/docs/llms.txt > Use this file to discover all available pages before exploring further. # Chat capabilities > The tools a chat agent can use inside a lane — read and edit files, run commands, drive Git, open PRs, and capture proof. A chat agent works like a teammate who has the lane checked out. ADE gives every agent a set of tools scoped to its role, so a regular lane chat can do real work — read and edit files, run commands, create lanes and PRs, pull in Linear context, and capture proof — without reaching the elevated control-plane tools reserved for the CTO. ## Universal tools Every agent gets these, in every session. > [card] Read files > > Range-aware file reads, plus `grep` and `glob` search across the worktree. Always allowed, even in read-only plan mode. > [card] Edit and write files > > Apply a single-file edit or create and replace a file. Each write produces a file-change event and a diff. Blocked in plan mode. > [card] Run commands > > Run shell commands for tests, builds, package managers, and scripts. Emits command events with output and exit code. Blocked in plan mode and gated by your approval in edit mode. > [card] Search the web > > Fetch a URL or run a web search when the task needs outside information. Always allowed. > [card] Track a task list > > Maintain a session todo list so multi-step work stays visible in the transcript. > [card] Ask for input > > Pause cleanly and ask you for a choice or clarification instead of guessing. The composer locks until you answer. ## Workflow tools Chat agents also get the tools that turn a conversation into shipped work. > [card] Create a lane > > Branch a fresh git worktree for a new piece of work, optionally stacked on the current lane. > [card] Open a PR > > Create a pull request from the lane's changes, with a generated title and body. > [card] Resolve PR issues > > When a chat is launched to fix a PR, it can refresh checks, re-run failed CI, reply to review threads, and resolve them. > [card] Capture proof > > Screenshot the environment and file the result to the proof drawer (macOS), or attach an existing image with `ade proof attach`. > [card] Report completion > > Persist a structured closeout — status, summary, artifacts, and any blocker — as a card in the transcript. > [card] Attach Linear > > When Linear is connected, a chat can read a linked issue's context; attaching an issue links it back to the lane and the next PR body. > [!NOTE] > > The exact set depends on the session: writes and commands are gated by the permission mode, screenshot capture needs a supported platform, and Linear tools appear only when the integration is connected. The CTO additionally gets operator tools (spawning chats, lanes/PRs/git/tests, durable memory, and Linear reads and updates) that regular chats never see. ## Provider-specific support Some capabilities depend on the provider runtime behind the chat. > [card] Codex goals > > Codex sessions can keep a visible goal that you set, update, complete, or clear from ADE without turning the goal into prompt text. > [card] Scheduled work > > Every chat runtime can schedule durable self-resume with `chat.createScheduledWork`; wake, cron, and background events stay inspectable in the transcript and Chat Info. > [card] Provider recovery > > Codex and Cursor runtime failures surface as structured recovery cards or errors instead of collapsing into silent transcript gaps. > [card] History handoff > > Claude and Codex sessions can be imported or forked into ADE when provider-native history is available, then continued as lane-scoped chats. Scheduled chat work resumes one chat at a turn boundary. For event-triggered, rule-based workflows across ADE, use [automations](https://ade-app.dev/docs/tools/automations.md). ## Run a command from a reply When a finished reply contains a shell code block, a **Run** button sits beside Copy (it appears on hover or focus). Run first shows the exact command, the machine it will run on, and the lane folder; nothing runs until you confirm. The command then runs in a new terminal in the chat's tools pane, on the chat's own machine, in the chat's lane folder. Pagers are turned off, so `git log` or `git diff` finishes instead of waiting at a prompt. When it finishes, the block shows the exit code and offers **Send output to the agent**, which sends the command, exit code and output as a normal message. Nothing is sent without that click. * `bash`, `sh`, `zsh` and `shell` blocks run through that shell, whatever your login shell is. A `console` block runs only its `$` or `%` prompt lines. * `powershell`, `pwsh` and `ps1` blocks run in Windows PowerShell on a Windows chat. * Run is disabled, with the reason, when the block's shell is not on the chat's machine (a bash block on Windows, a PowerShell block on a Mac), for `cmd` and `bat` blocks, for scripts too long for a command line, for blocks with hidden control characters, and in a personal chat, an archived lane, or on an offline machine. * Run never appears while the reply is still streaming, or on your own messages. ## See what an agent can reach Open the tools panel from the composer to see what a session has available and which actions need your approval before they run. ## The right agent for the tool Capabilities are shared across providers, but the model still matters. Pick the provider and model that suits the task when you start the session — a fast model for a small fix, a stronger one for a gnarly refactor — and switch within the same family mid-session. ## Good chat requests Specific, verifiable asks get better results than open-ended cleanup. ```text theme={null} Find why the billing test fails and fix the smallest underlying bug. ``` ```text theme={null} Update the PR description from the current diff and include test evidence. ``` ```text theme={null} Read the auth module and explain where the redirect state is stored. ``` ## Review before trusting Agent chat is powerful because it changes real files. Treat the result like any contribution: inspect the diff in **Files**, run the relevant checks, and read the closeout summary before you commit or merge. > [card] [Context and artifacts](https://ade-app.dev/docs/chat/context.md) > > What's in context, and how to keep a session focused. > [card] [Files & diffs](https://ade-app.dev/docs/tools/files-editor.md) > > Read, stage, and resolve the changes an agent makes. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.