### Run ftui-harness examples

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/tutorials/agent-harness.md

Execute the minimal and streaming examples for the ftui-harness.

```bash
cargo run -p ftui-harness --example minimal

```

```bash
cargo run -p ftui-harness --example streaming

```

--------------------------------

### Run Harness Examples

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Execute harness examples to test and reference behavior. Use these to verify specific functionalities.

```bash
cargo run -p ftui-harness --example minimal
```

```bash
cargo run -p ftui-harness --example streaming
```

--------------------------------

### Run Streaming Example

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/getting-started.md

Command to run the streaming example from the ftui-harness crate.

```bash
cargo run -p ftui-harness --example streaming
```

--------------------------------

### Install FrankenTUI from Source Tarball

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Use this method to quickly install FrankenTUI by downloading and building the source tarball. Ensure you have Rust and Cargo installed.

```bash
curl -fsSL https://codeload.github.com/Dicklesworthstone/frankentui/tar.gz/main | tar -xz
cd frankentui-main
cargo build --release
```

--------------------------------

### bv Scoping and Filtering Examples

Source: https://github.com/dicklesworthstone/frankentui/blob/main/AGENTS.md

Examples demonstrating how to scope bv analysis to specific labels, historical points, or pre-filtered recipes.

```bash
bv --robot-plan --label backend              # Scope to label's subgraph
```

```bash
bv --robot-insights --as-of HEAD~30          # Historical point-in-time
```

```bash
bv --recipe actionable --robot-plan          # Pre-filter: ready to work
```

```bash
bv --recipe high-impact --robot-triage       # Pre-filter: top PageRank
```

```bash
bv --robot-triage --robot-triage-by-track    # Group by parallel work streams
```

```bash
bv --robot-triage --robot-triage-by-label    # Group by domain
```

--------------------------------

### Install Nightly Rust Toolchain

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

FrankenTUI requires the nightly Rust compiler. Install it using rustup.

```bash
rustup toolchain install nightly
```

--------------------------------

### Minimal FrankenTUI Application Example

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

This example demonstrates a basic FrankenTUI application that displays a tick counter and quits when 'q' is pressed. It requires the `ftui_core`, `ftui_render`, `ftui_runtime`, and `ftui_widgets` crates.

```rust
use ftui_core::event::Event;
use ftui_core::geometry::Rect;
use ftui_render::frame::Frame;
use ftui_runtime::{App, Cmd, Model, ScreenMode};
use ftui_widgets::paragraph::Paragraph;

struct TickApp {
    ticks: u64,
}

#[derive(Debug, Clone)]
enum Msg {
    Tick,
    Quit,
}

impl From<Event> for Msg {
    fn from(e: Event) -> Self {
        match e {
            Event::Key(k) if k.is_char('q') => Msg::Quit,
            _ => Msg::Tick,
        }
    }
}

impl Model for TickApp {
    type Message = Msg;

    fn update(&mut self, msg: Msg) -> Cmd<Msg> {
        match msg {
            Msg::Tick => {
                self.ticks += 1;
                Cmd::none()
            }
            Msg::Quit => Cmd::quit(),
        }
    }

    fn view(&self, frame: &mut Frame) {
        let text = format!("Ticks: {}  (press 'q' to quit)", self.ticks);
        let area = Rect::new(0, 0, frame.width(), 1);
        Paragraph::new(text).render(area, frame);
    }
}

fn main() -> std::io::Result<()> {
    App::new(TickApp { ticks: 0 })
        .screen_mode(ScreenMode::Inline { ui_height: 1 })
        .run()
}
```

--------------------------------

### Run Minimal Inline App

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/getting-started.md

Command to compile and run the minimal inline application example.

```bash
cargo run
```

--------------------------------

### Preset Backdrop Configurations

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/concepts/visual-fx.md

Provides examples of using the `subtle()` and `vibrant()` presets for common backdrop configurations, offering quick setup for legibility-first or visually impactful designs.

```rust
// Subtle: 15% opacity, no scrim (prioritizes legibility)
Backdrop::new(Box::new(fx), theme).subtle().over(&child);

// Vibrant: 50% opacity, vignette scrim (visual impact)
Backdrop::new(Box::new(fx), theme).vibrant().over(&child);
```

--------------------------------

### Canonical Entrypoint for FrankenTUI Application

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/planning/plan-to-create-frankentui-codex.md

This example demonstrates the primary way to create and run a FrankenTUI application. It shows the basic structure of a `Model` implementation and how to initialize and launch the `App` with specific screen mode settings.

```APIDOC
## Canonical entrypoint
```rust
use ftui::{App, Cmd, Event, Frame, Model, ScreenMode, Result};

struct State { count: u64 }

impl Model for State {
    type Message = Event;

    fn update(&mut self, msg: Event) -> Cmd<Event> {
        match msg {
            Event::Key(k) if k.is_char('q') => Cmd::quit(),
            Event::Key(k) if k.is_char('+') => {
                self.count += 1;
                Cmd::none()
            }
            _ => Cmd::none(),
        }
    }

    fn view(&self, frame: &mut Frame) {
        // draw into frame
    }
}

fn main() -> Result<()> {
    App::new(State { count: 0 })
        .screen_mode(ScreenMode::Inline { ui_height: 6 })
        .run()
}
```
```

--------------------------------

### Configure Evidence Logging in Rust

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Example of configuring and enabling structured evidence logs for FrankenTUI. This setup directs logs to a file named 'evidence.jsonl' and flushes them immediately after writing.

```rust
use ftui_runtime::{EvidenceSinkConfig, EvidenceSinkDestination, Program, ProgramConfig};

let config = ProgramConfig::default().with_evidence_sink(
    EvidenceSinkConfig::enabled_file("evidence.jsonl")
        .with_destination(EvidenceSinkDestination::file("evidence.jsonl"))
        .with_flush_on_write(true),
);

let mut program = Program::with_config(model, config)?;
program.run()?;
```

--------------------------------

### Install cargo-llvm-cov

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/testing/coverage-playbook.md

Installs the `cargo-llvm-cov` tool if it is not already present.

```bash
cargo install cargo-llvm-cov
```

--------------------------------

### Clone and Run Demo Showcase with Git

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Clone the FrankenTUI repository using git, navigate into the project directory, and then run the demo showcase.

```bash
git clone https://github.com/dicklesworthstone/frankentui.git
cd frankentui
cargo run -p ftui-demo-showcase
```

--------------------------------

### Command Examples

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Demonstrates various ways to create and use Commands for handling side effects and asynchronous operations.

```rust
Cmd::none()                    // No side effect
Cmd::perform(future, mapper)   // Async operation → Message
Cmd::quit()                    // Exit program
Cmd::batch(vec![...])          // Multiple commands
```

--------------------------------

### Subscription Examples

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Shows how to define subscriptions for declarative, long-running event sources like timers and file watchers.

```rust
fn subscriptions(&self) -> Vec<Box<dyn Subscription<Message>>> {
    vec![
        tick_every(Duration::from_millis(16)),   // 60fps timer
        file_watcher("/path/to/watch"),          // FS events
    ]
}
```

--------------------------------

### Run Demo with Pre-recorded Scenario

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/macro-recorder.md

Executes the FrankenTUI demo showcase using a pre-recorded macro scenario specified by the FTUI_DEMO_MACRO environment variable.

```bash
# Run demo with a pre-recorded scenario
FTUI_DEMO_MACRO=scenarios/quick-tour.json cargo run -p ftui-demo-showcase
```

--------------------------------

### E2E Event JSONL Example Fixture

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/testing/e2e-summary-schema.md

This file serves as a handy example fixture for local verification of E2E Event JSONL.

```json
tests/e2e/lib/e2e_jsonl_examples.jsonl
```

--------------------------------

### Run Performance HUD Demo Showcase

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/performance-hud.md

Launches the demo showcase and directly navigates to the Performance screen. Toggle the HUD with Ctrl+P.

```bash
cargo run -p ftui-demo-showcase -- --screen=10
```

--------------------------------

### Confidence Ledger Entry Example

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/spec/telemetry-perf.md

An example of a confidence ledger entry in JSONL format, detailing benchmark results and confidence metrics.

```json
{
  "run_id": "20260208T202350-2604979",
  "benchmark": "web/frame_harness_stats/heavy_50pct/80x24",
  "status": "PASS",
  "actual_ns": 10000,
  "budget_ns": 30000,
  "ci_low_ns": 9900,
  "ci_high_ns": 10100,
  "ci_width_ns": 200,
  "relative_ci_width": 0.006667,
  "sigma_ns": 150.0,
  "variance_ns2": 22500.0,
  "posterior_prob_regression": 0.0,
  "e_value": 1.0,
  "bayes_factor_10": 0.196116,
  "decision": "allow",
  "confidence_hint": "pass",
  "loss_matrix": { "false_positive": 1, "false_negative": 5 },
  "hardware": {
    "os": "Linux",
    "arch": "x86_64",
    "cpu_model": "AMD EPYC 7B13",
    "cpu_cores": 16
  }
}
```

--------------------------------

### Download and Run Demo Showcase with Curl

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Download the FrankenTUI source code using curl, extract it, navigate to the directory, and then run the demo showcase.

```bash
# Download source with curl (no installer yet)
curl -fsSL https://codeload.github.com/dicklesworthstone/frankentui/tar.gz/main | tar -xz
cd frankentui-main

# Run the demo showcase (primary way to see what FrankenTUI can do)
cargo run -p ftui-demo-showcase
```

--------------------------------

### Diff Decision Event Example

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/spec/telemetry-events.md

Example JSON log for a 'diff_decision' event. This event captures details about the diffing strategy and its performance metrics.

```json
{"event":"diff_decision","run_id":"diff-4242","event_idx":12,"strategy":"DirtyRows","cost_full":4800.000000,"cost_dirty":1200.000000,"cost_redraw":0.000000,"posterior_mean":0.036000,"posterior_variance":0.000340,"alpha":3.500000,"beta":92.500000,"dirty_rows":10,"total_rows":40,"total_cells":4800,"span_count":6,"span_coverage_pct":12.500000,"max_span_len":18,"fallback_reason":"none","scan_cost_estimate":600,"tile_used":true,"tile_fallback":"none","tile_w":16,"tile_h":16,"tile_size":256,"tiles_x":5,"tiles_y":3,"dirty_tiles":2,"dirty_tile_count":2,"dirty_cells":40,"dirty_tile_ratio":0.133333,"dirty_cell_ratio":0.008333,"scanned_tiles":2,"skipped_tiles":13,"skipped_tile_count":13,"tile_scan_cells_estimate":512,"sat_build_cost_est":4800,"bayesian_enabled":true,"dirty_rows_enabled":true}
```

--------------------------------

### FrankenTUI Environment Variables Configuration

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Example .env file showing available environment variables for configuring FrankenTUI's harness behavior, including screen mode, UI height, view selection, mouse and focus support, and logging options.

```bash
# .env (example)
FTUI_HARNESS_SCREEN_MODE=inline   # inline | alt
FTUI_HARNESS_UI_HEIGHT=12         # rows reserved for UI
FTUI_HARNESS_VIEW=layout-grid     # view selector
FTUI_HARNESS_ENABLE_MOUSE=true
FTUI_HARNESS_ENABLE_FOCUS=true
FTUI_HARNESS_LOG_LINES=25
FTUI_HARNESS_LOG_MARKUP=true
FTUI_HARNESS_LOG_FILE=/path/to/log.txt
FTUI_HARNESS_EXIT_AFTER_MS=0      # 0 disables auto-exit
```

--------------------------------

### Example Pane Monitor Explanation

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/perf/pane_assumption_monitors.md

An example of an operator-readable explanation for a violated pane monitor assumption, detailing the observed value, budget, and the impact.

```text
> Checkpointed replay walks 40 step(s), 2.5x the 16-step checkpoint interval
> (checkpoint_hit=false) — undo/redo will feel sluggish; the checkpoint-spacing
> assumption is violated.
```

--------------------------------

### Install Nightly Rust Compiler

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/getting-started.md

If you encounter errors related to the nightly compiler, install it using rustup. This ensures compatibility with features required by FrankenTUI.

```bash
rustup toolchain install nightly
```

--------------------------------

### Canonical Entrypoint: ftui::App Example

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/planning/plan-to-create-frankentui-opus.md

Demonstrates the basic structure of an application using `ftui::App`. It defines a `State` model and implements the `Model` trait to handle events and render UI. This serves as the default path for inline, scrollback, and stable UI regions.

```rust
use ftui::{App, Cmd, Event, Frame, Model, ScreenMode, Result};

struct State { count: u64 }

impl Model for State {
    type Message = Event;

    fn update(&mut self, msg: Event) -> Cmd<Event> {
        match msg {
            Event::Key(k) if k.is_char('q') => Cmd::quit(),
            Event::Key(k) if k.is_char('+') => { self.count += 1; Cmd::none() } 
            _ => Cmd::none(),
        }
    }

    fn view(&self, frame: &mut Frame) {
        // draw into frame.buffer using text/style/layout helpers
    }
}

fn main() -> Result<()> {
    App::new(State { count: 0 })
        .screen_mode(ScreenMode::Inline { ui_height: 6 })
        .run()
}
```

--------------------------------

### JSONL Log Format Example

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/spec/wasm-showcase-runner-contract.md

Example of host-side JSONL log lines generated from WASM runner data. These are used for end-to-end testing and CI.

```jsonl
{"event":"step","frame_idx":1,"ts_ns":16000000,"rendered":true,"events_processed":3}
{"event":"patch_stats","frame_idx":1,"dirty_cells":42,"patch_count":3,"bytes_uploaded":672}
{"event":"frame","frame_idx":1,"patch_hash":"fnv1a64:a1b2c3d4e5f6a7b8"}
```

--------------------------------

### BOCPD Event Example

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/spec/telemetry-events.md

Example JSON log for a 'bocpd' event, which is schema-versioned. This event captures Bayesian Online Change Point Detection information.

```json
{"schema_version":"bocpd-v1","event":"bocpd","p_burst":0.7321,"log_bf":1.204,"obs_ms":18.0,"regime":"burst","ll_steady":0.001234,"ll_burst":0.056789,"runlen_mean":12.4,"runlen_var":9.100,"runlen_mode":9,"runlen_p95":21,"runlen_tail":0.0420,"delay_ms":40,"forced_deadline":false,"n_obs":64}
```

--------------------------------

### Run FrankenTUI Demo Showcase

Source: https://github.com/dicklesworthstone/frankentui/blob/main/README.md

Execute the primary demo showcase for FrankenTUI. You can also specify a particular view to load using the FTUI_HARNESS_VIEW environment variable.

```bash
# Demo showcase (primary)
cargo run -p ftui-demo-showcase
```

```bash
# Pick a specific demo view
FTUI_HARNESS_VIEW=dashboard cargo run -p ftui-demo-showcase
FTUI_HARNESS_VIEW=visual_effects cargo run -p ftui-demo-showcase
```

--------------------------------

### Input Event Record Example

Source: https://github.com/dicklesworthstone/frankentui/blob/main/docs/spec/frankenterm-golden-trace-format.md

Input records capture effective input semantics for replay. For web/remote, prefer post-normalization. This example shows a key input.

```json
{"schema_version":"golden-trace-v1","event":"input","ts_ns":16000000,"kind":"key", ...}
```