donaldfilimon/abi / .agents/skills/opencode/plugins/tui
tui
VerifiedTUI/dashboard OpenCode plugin. Launch interactive agent REPL, diagnostics dashboard panes, and slash commands via real abi CLI paths. There is no /abi-superpower-tui binary.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 3.4 KB |
SKILL.md
View on GitHub---
name: tui
description: TUI/dashboard OpenCode plugin. Launch interactive agent REPL, diagnostics dashboard panes, and slash commands via real abi CLI paths. There is no /abi-superpower-tui binary.
superpower:
command: "execute"
parameters:
- name: "action"
type: "string"
enum: ["repl", "dashboard", "pane", "once", "json"]
description: "TUI action"
- name: "pane"
type: "string"
description: "Dashboard pane: system, plugins, storage, wdbx, scheduler, memory"
- name: "compact"
type: "boolean"
description: "Render only selected pane"
---
# TUI Superpower Plugin
Core TUI capabilities for OpenCode within the ABI framework. There is **no**
`/abi-superpower-tui` binary or slash command — map actions to the real CLI
paths below (or the `run-tui` / `dashboard-smoke` drivers).
## Capabilities
- TUI subsystem integration (dashboard, agent REPL, line editor)
- Plugin framework registration
- Runtime lifecycle management
- Configuration and settings management
- Status monitoring and reporting
## Integration Points
- ABI's TUI subsystem (`abi tui` / `abi dashboard` / `abi agent tui`)
- OpenCode plugin framework integration
- Runtime lifecycle management
- Configuration and settings management
## Actions
### repl
Launch the agent REPL (slash commands live here):
```bash
./target/debug/abi agent tui
```
### dashboard
Show the diagnostics dashboard (`abi tui` is an alias of `abi dashboard`):
```bash
./target/debug/abi dashboard --pane system
./target/debug/abi dashboard --pane plugins --once --json
./target/debug/abi tui --compact --pane scheduler
```
Interactive pty smoke (tmux driver):
```bash
.agents/skills/run-tui/tui.sh # drives `abi dashboard`
.agents/skills/run-tui/tui.sh tui # drives `abi tui`
```
Headless one-shot smoke:
```bash
.agents/skills/dashboard-smoke/dashboard.sh
```
### pane
Select the initial diagnostics pane with `--pane` (system, plugins, storage/wdbx,
scheduler, memory, or 1–5). In the interactive refresh loop, switch panes with
hotkeys / Tab — there is no separate `pane --focus` CLI.
```bash
./target/debug/abi dashboard --list-panes
./target/debug/abi dashboard --pane memory
```
## Slash Commands (in `abi agent tui` REPL)
- `/help` - Show the REPL command surface
- `/model <id>` - Select a model
- `/profile` - Report adaptive profile state
- `/sea [on|off|status|toggle]` - Change session-local preference only
- `/status`, `/context`, `/history`, `/reset`, `/features`, `/clear`, `/quit`
The REPL does not implement `/open`, `/diff`, `/commit`, `/learn`, `/save`, or
`/load`. Actual evidence recall is the separate `abi complete --learn` path.
## Implementation
Maps to:
- `crates/abi-cli/src/repl.rs` - REPL commands and session state
- `crates/abi-cli/src/repl_editor.rs` - bounded line editor and raw input loop
- `crates/abi-cli/src/dashboard/` - Split-pane dashboard (`abi tui` / `abi dashboard`)
- `crates/abi-cli/src/terminal.rs` - raw-mode guard and key decoder
- `.agents/skills/run-tui/tui.sh` - Interactive pty driver
- `.agents/skills/dashboard-smoke/dashboard.sh` - Headless one-shot smoke
## Runtime Boundary
The Rust CLI always includes these surfaces. Raw mode requires a Unix TTY;
redirected input and dashboard `--once`/`--json` use deterministic fallbacks.
Dashboard panes support bounded keyboard navigation and primary-button SGR
mouse selection, with capture cleanup on every exit path.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |