VCode home

Read the transcript

Agents: claude, codex, opencode · On: phone, desktop

The middle of the screen (#thread) is the conversation: your messages, the agent's replies, its thinking, every tool it ran and every file it edited, in the order they happened.

Why you would use it

This is where you follow what the agent is doing and check what it did. On a phone it is also where you go back over a long turn (a command's output, the lines of an edit) without opening a terminal.

How to use it

  1. Scroll the transcript (#thread). It opens at the newest message.

    A thread opened at its newest cards: the end of a reply, a SYS compaction card, a YOU message, a QUEUED message and the usage-limit card

  2. Tap a card that is clamped to open it in full; tap it again to fold it back.

    The top of the same thread: a YOU card, a BASH tool card with its output, and an EDIT card showing the first lines of its patch

  3. Tap ▴ N earlier at the top to load more of the history.

  4. Tap the pencil on one of your own messages to pull that prompt back into the composer, or the copy icon to copy the text.

  5. Tap the diff icon on an Edit card to open the whole turn's diff in the diff sheet, at that file.

What you see

Every card is headed by an icon and a word, with no box around them. Each card for something the agent did (thinking, a tool call, an edit, a Sys note, its reply) has a grey icon in a left gutter, with its word beside it. A thin line joins those icons from the first step of a turn down to the reply, so one turn reads as one run. A message you send breaks the line. A step that failed has a red icon.

Tag Card
a green person icon plus You a message you sent
VCode's mark plus Agent Commander, or a tab's agent logo plus the tab's title a message that tab sent with v-code tab send or tab new --prompt. The name is underlined: tap it to open that tab
a speech bubble plus Report a tab's report in an older Agent Commander conversation, from before tabs stopped reporting
the agent's logo plus its name and model a reply, e.g. "Claude Opus 5"
Thinking (a thought bubble) reasoning. On OpenCode, a reply that opens with the model's own <think> or <thinking> block shows that block here and the rest as the reply
Read Write Edit Bash Search Web Subagent Plan Wakeup Tools Message Stop Think Tool a tool call, named for what the tool does, each with its own icon: a terminal for Bash, a page for Read, a magnifier for Search, a globe for Web, a network for Subagent, a square pen for Edit. A tool not on that list shows the whole words of its name that fit in ten characters, with a wrench; an MCP tool shows its own name without the mcp__server__ prefix
Edit plus a file path and +N −N a file the agent changed
Sys VCode talking, e.g. an agent's notice. An info icon plus System, then Restart or Update v0.0.613 → v0.0.615, is the message VCode sent the agent to carry on a turn a restart cut short
the orange ! glyph plus Needs you an approval request
an amber clock plus Queued, or an amber upload arrow plus Outbox a message waiting to be sent: queued or held offline

Beside the tag: the time for a message, the command for a shell tool (coloured as shell), the file path for a read or an edit. On a Read or Write card the file name is underlined: tap it to open the file in the viewer, at the first line the agent read. The rest of the head still folds the card. On the right: how long the step took (1.4s), counted from the end of the step before it, so the time the agent spent writing the command counts too; … while it is still running, or the added and removed line counts on an edit.

A tool card that is still running says running…. An edit card shows the first 8 lines of its patch until you open it.

Long cards are clamped: tool output over 8 lines or 700 characters, diffs over 8 lines or 1200 characters, anything else over 22 lines or 1600 characters. The last card in the thread is exempt when it is the reply itself, so the answer is never hidden behind a tap. Tapping a link, inline code or a code block inside a card does not fold the card. A table is part of the text: tapping it opens or folds the card like the prose around it.

A message from another tab has no pencil, only the copy icon: nobody typed it here. Its name is the tab's title now; a tab this app no longer shows keeps the title it had when it sent, and tapping it says That tab is closed.

A tab's transcript with a message from the Agent Commander, under VCode's mark, and one from another tab, under that tab's agent logo and title

A SYS Restart card is one line until you tap it; open, it shows the text the agent was sent.

A turn that changed files gets a footer under its last reply: a file icon, N files and the totals. Tapping it opens the list of files.

Cards that inform nobody are hidden: empty wait and ToolSearch polls, TaskCreate and TaskUpdate calls, one-word subagent acknowledgements, and "The file … has been updated successfully" under an Edit or Write. Errors and running calls are never hidden.

The cog's Transcript group sets how much of each thinking, tool, edit and system card shows: in full, on one line, grouped or hidden. See stream-detail-levels.

Limits and known gaps