hyperhive/docs/terminal-rendering.md

Per-agent terminal: row taxonomy (as built)

Snapshot of how the per-agent web UI's live pane renders each event kind today. Source of truth lives in frontend/packages/agent/src/app.js (renderStream, fmtToolUse, renderRichToolUse, renderToolResult, renderTaskEvent, mdNode, detailsOpenMd, fmtArgsGeneric) + frontend/packages/shared/src/terminal.css (the shared .live .<class> styling) + the marked npm package (markdown).

Layout contract

Every row — flat <div class="row …"> and expandable <details class="row …"> alike — shares one prefix column. The mechanism is padding-left + negative text-indent on .live .row: the row's first inline box gets pulled back into the column at ~0.5em, and wrapped continuation lines hang under the body, not under the glyph.

Rows that carry an icon (the per-tool emoji, 🧠/💭 thinking, etc.) pass it as the icon argument to row() / details() / detailsDiff(), which puts it in a fixed-width .row-glyph cell (display: inline-block; width: 1.4em) rather than as a bare first character. The constant cell width means every icon's left edge lines up in the column regardless of the glyph's rendered width (emoji differ; some carry a variation selector) — a flat row's 🧠 and a details summary's 🖥️ align. Rows with a plain single-char glyph (◆ · ! ←) still pass it inline; it lands at the same ~0.5em left edge.

<details> summaries inherit those metrics. The icon (when present) sits in the .row-glyph cell; the summary text lives in a .summary-text span and the disclosure caret (▸ / ▾) is supplied by CSS .summary-text::before so it leads the text, not the icon — a leading caret on the icon would push it out of the shared column. Icon-less summaries have no .row-glyph, so the caret falls into the prefix column like the old directional glyph. The summary text carries no → / ←; the row colour (cyan = outbound, muted = inbound) carries the direction.

Child blocks inside a row (the .md markdown wrapper, an inner <details>) get text-indent: 0 so their content lays out from the body column instead of inheriting the parent's negative pull.

Row taxonomy

CSS class	Prefix glyph	Color	Triggered by	Source
.turn-start	◆ TURN ← <from>	amber, left rule	LiveEvent::TurnStart	harness wake
.turn-body	(child div under turn-start)	fg	same	the wake-prompt body
.turn-end-ok	✅ turn ok	green, left rule	LiveEvent::TurnEnd { ok: true }	harness
.turn-end-fail	❌ turn fail — note	red, left rule	LiveEvent::TurnEnd { ok: false }	harness
.turn-time	· HH:MM:SS on turn-start; · HH:MM:SS · <dur> on turn-end (child span)	muted, smaller	per-event ts (unix seconds) on the live frame + history row	harness
.text	(no prefix; markdown body)	fg	claude assistant.content[].text	stream-json
.thinking	💭 thinking …	muted, italic	claude assistant.content[].thinking	stream-json
.tool-use (flat)	<icon> Name args…	cyan	tool_use w/o rich renderer; <icon> from toolIcon(name) (📤 send · 📥 recv · ❓ ask · ⏰ remind · 🏷️ set_status · 🪢 loose-ends · 🖥️ bash · 💬 matrix · 📦 request_* · ⏱️ schedule · 🔧 default)	stream-json
.tool-use <details>	💾/✏️ Write/Edit <path> · +N (no →)	cyan, body is +/- diff	renderRichToolUse Write/Edit	stream-json
.tool-use <details open>	📤 send → to · NL, ❓ ask → to, ✍️ answer #id	cyan, body is markdown	rich renderer for send / ask / answer	stream-json
.tool-use .ask-answer-inline-slot	(sub-block under ask → operator)	inherits row	inline answer form bound by reconcileAskBinds to the loose-end	rich renderer
.tool-result (flat)	← <txt>	muted	short tool_result (≤120c, non-recv)	stream-json
.tool-result-block <details>	Nl · headline	muted, body is text	long generic tool_result	stream-json
.tool-result-block <details open>	recv ← <txt>	muted, body is markdown	tool_result correlated to a prior recv tool_use via id	stream-json
.tool-result.error (flat)	✗ <msg>	red	tool_result with is_error: true (≤120c); <tool_use_error> wrapper stripped	stream-json
.tool-result-block.error <details>	Nl · headline	red, body is text	long error tool_result (is_error: true); wrapper stripped	stream-json
.tool-use	⌁ task <id> started · <desc> [type]	cyan	claude Task-tool subagent start (dead path — Task omitted from agent allow-list)	renderTaskEvent
.turn-end-ok / .turn-end-fail / .tool-result	⌁ task <id> ✓/✗/◌ <status> · <desc> · → <output_file>	green / red / muted	claude Task-tool result (dead path for agents)	renderTaskEvent
.note	· ⚙ plugin install · loading… or ✓ done	muted	system/plugin_install (status = started/completed)	stream-json
.note	· ⚙ commands changed · N available (expandable list of /name entries)	muted	system/commands_changed (slash-command set updated, usually post-plugin_install)	stream-json
.note	· ⚙ compact · <trigger> · <pre>→<post> tokens · <dur>	muted	system/compact_boundary (compaction completed; metadata includes pre/post token counts, duration, trigger)	stream-json
.note	· ⚙ <subtype>	muted	other system subtypes (context_window_exceeded, etc.)	stream-json catch-all
.note	· <text>	muted	harness chatter	LiveEvent::Note
.note.stderr	! stderr: <line>	amber/orange	stderr lines off claude	LiveEvent::Note (text starts stderr:)
.note.op	· operator: <text>	mauve italic	operator-initiated notes (/cancel, /compact, /model, new-session)	LiveEvent::Note (text starts operator:)
.sys	! {json…}	amber/orange	catch-all for stream shapes renderStream didn't classify	catch-all
Banner shimmer	mauve	turn in flight (ref-counted)	—	setBannerActive

The .turn-time span is appended to the turn-start / turn-end rows from the event's ts (unix seconds), which the backend serializes as a flattened sibling of kind on both the live SSE frame and each history row — so the same renderer path stamps live tail and replayed scrollback identically. Turn-end also shows the elapsed duration (end − start), paired against the most recent open turn-start. The read is guarded on a numeric ts: if a frame omits it the rows render without the time suffix, so the terminal degrades cleanly against older event shapes.

Renderer dispatch

renderStream(v, api) walks each stream-json line:

1. Drops system/init, rate_limit_event, result (noise / handled elsewhere — result powers the cost badge). 1a. system/thinking_tokens (claude streams a running estimated_tokens counter while thinking — many per turn) → collapses into a single 🧠 thinking … ~N tokens .note row that updates in place. Consecutive ticks reuse the row only while it's still the last one rendered (nextElementSibling == null); any other event after it makes the next tick start a fresh row. Avoids a note-per-tick scrollback flood. 1b. system/plugin_install → muted note ⚙ plugin install · loading… (on started) or ⚙ plugin install · ✓ done (on completed). Emitted in pairs: started fires before the plugin loads, completed fires when it's ready. The uuid links the pair. 1c. system/commands_changed → collapsible .note details row showing the new slash-command count (⚙ commands changed · N available). Expanding reveals each /name and its aliases. Fires after plugin_install when a plugin registers new commands. 1d. system/compact_boundary → muted note showing compaction summary: ⚙ compact · <trigger> · <pre>→<post> tokens · <dur>. Fields are guarded individually — a missing field is silently omitted. Trigger is "manual" (operator /compact) or "auto". 1e. Other system/ subtypes → muted note ⚙ <subtype>.
2. subtype == "task_started" | "task_notification" → renderTaskEvent (subagent activity gets the ⌁ glyph).
3. type == "assistant" → walk message.content[]:
   • text → .text row with a markdown body via mdNode.
   • thinking → .thinking row.
   • tool_use → record id → name in toolNameById, try renderRichToolUse (Write/Edit/send/ask/answer get custom renderings); on miss fall through to a flat .tool-use row with fmtToolUse → fmtArgsGeneric. fmtToolUse surfaces the salient arg per built-in tool — e.g. recv shows wait <N>s / max <N> when set (bare recv() otherwise), Bash flags [bg] for run_in_background commands.
4. type == "user" → walk message.content[] for tool_result; renderToolResult correlates via tool_use_id → toolNameById to default-open recv results with a markdown body, else short = flat / long = collapsed details.
5. Unrecognised shape → .sys row (amber, ! glyph).

Markdown

mdNode(text) wraps marked.parse(text) (the marked v4.x npm dep, bundled by esbuild into the page's app.js) in a <div class="md">. CSS in terminal.css scopes paragraph / code / list / blockquote / link styling under .live .row .md so the markdown body doesn't bleed into the row's own text-indent. Falls back to plain text if marked didn't load. Applied to text rows and to send / ask / answer / recv message bodies.

Extra-MCP tools

fmtArgsGeneric(name, input) is the fallback when a tool isn't in the built-in fmtToolUse switch:

• single string field → name k: "v"
• single number/bool field → name k: v
• multi-field → first 4 pairs trimmed to k: "v" / k: [N] / k: {…} with a …+N overflow

This keeps mcp__matrix__send_message and similar from dumping raw JSON.

Inline ask-operator answer

When an agent calls mcp__hyperhive__ask with to == "operator" (default), the rich tool-use renderer mounts an empty <div class="ask-answer-inline-slot"> inside the row's expanded body, then enqueues a loose-ends refresh. reconcileAskBinds() runs on every loose-ends refresh, matches each waiting slot against pending operator-bound questions by question text, and injects an inline .answer-form (textarea + send button bound to /answer-question/<id> on the host dashboard) into the matching slot. When a question subsequently leaves the pending list (answered, cancelled by asker, or TTL-expired), the same reconciler replaces the form with a struck-through [resolved] tag so the scrollback reflects the closed state. The label is neutral because /api/loose-ends only carries pending state — full resolution detail is visible via the question's history in the side panel.

Lets the operator answer mid-flow without context-switching to the loose-ends side panel or the dashboard tab. Side panel + dashboard forms remain — they're the same buildAnswerForm factory, three mount points for the same POST.

Dashboard side (not covered here)

The main dashboard's message-flow pane is a different shape: broker messages render as .msgrow grid lines (ts / arrow / from / → / to / body) with their own styling. .live .msgrow explicitly resets text-indent: 0 so the per-agent terminal's hanging-indent metrics don't leak into the flex-grid broker rows.