feat: replace --ui monitor entrypoint with dash backend foundations

.codex/skills/qa-night-shift/SKILL.md

@@ -131,6 +131,8 @@ Typical flow:
    resume attempt when the run was interrupted
 10. use `night-shift resolve` or `night-shift resume` only if the run actually
    requires it
+11. use `night-shift dash` when the QA pass needs browser-visible bootstrap,
+    SSE, command, audit, or raw artifact validation
 
 For review-driven investigations, replace steps 3-4 with:
 
@@ -143,6 +145,8 @@ In review-driven runs, pay attention to repo-state evidence:
 
 - the stored open-PR snapshot captured during planning
 - whether `status`, `report`, or the dashboard show repo-state drift
+- whether `night-shift dash` bootstrap and SSE payloads stay aligned with the
+  journal-backed repo-state snapshot
 - whether `night-shift report` shows the actionable/impacted subtree and
   replacement lineage, while the persisted `report.md` remains readable
   without live GitHub refresh
@@ -175,6 +179,8 @@ In review-driven runs, pay attention to repo-state evidence:
   failures with usable artifact paths
 - whether `status`, `report`, and the dashboard agree on the confidence posture
   and its reasons
+- whether Dash command handlers mutate the same durable run state as the CLI,
+  without scraping command output
 - whether `provenance` records the expected prompt paths, payload artifacts,
   verification evidence, worktree paths, and PR linkage
 - whether `doctor` classifies interrupted tasks as `safe_to_resume`,

README.md

@@ -21,7 +21,7 @@ Night Shift already has working support for:
 - provider adapters for Codex CLI and Cursor Agent
 - local verification before pull request delivery
 - review-loop ingestion for open Night Shift pull requests
-- a local monitor-only dashboard via `start --ui` and `resume --ui`
+- a localhost-only Dash control surface via `night-shift dash`
 
 The current operator flow is:
 
@@ -126,10 +126,10 @@ night-shift resume --explain
 night-shift resume
 ```
 
-If you want the local dashboard while a run is active:
+If you want the local dashboard:
 
 ```sh
-night-shift start --ui
+night-shift dash
 ```
 
 ## Source Development

docs/getting-started.md

@@ -100,11 +100,11 @@ Kick off the most recent pending run:
 night-shift start
 ```
 
-You can also target a specific run or open the local dashboard:
+You can also target a specific run or open Dash:
 
 ```sh
 night-shift start --run run-123
-night-shift start --ui
+night-shift dash
 ```
 
 `start` is execution-only. It uses the execution agent, environment, and brief
@@ -154,7 +154,6 @@ If a run was interrupted, resume from the saved journal:
 night-shift doctor
 night-shift resume --explain
 night-shift resume
-night-shift resume --ui
 ```
 
 `doctor` is the dry recovery pass. It classifies each task as

docs/providers-and-delivery.md

@@ -111,23 +111,24 @@ leave that disabled and still use the PR-body overlay.
 
 ## Dashboard
 
-The local dashboard is intentionally narrow in scope. It binds to `127.0.0.1`,
-prefers port `8787`, and serves a monitor-only UI for:
+The local dashboard now uses `night-shift dash`. It binds to `127.0.0.1` and
+serves a Dash backend for:
 
-- run history
-- summary metadata for the selected run
-- repo-state summary for review-driven runs, including snapshot time and drift
-- task status
-- event timeline
-- report content
+- structured bootstrap state for the current repository
+- SSE-first live updates
+- browser command handlers for `init`, `plan`, `plan --from-reviews`,
+  `resolve`, `start`, and `resume`
+- audit and raw artifact routes for reports, provenance, logs, payloads, and
+  runtime identity files
 
-There are no browser-side mutation controls.
+The browser surface reuses the same in-process usecases as the CLI rather than
+shelling out to `night-shift` subprocesses.
 
 ## Demo Mode
 
 `night-shift --demo` exercises a fixture-backed flow and prints a compact proof
 summary. The headless demo validates `plan`, `start`, `status`, and `report`.
-The UI demo validates the local dashboard payload as well.
+The UI demo validates Dash bootstrap plus browser-command execution as well.
 
 Demo artifacts live under:
 

docs/run-lifecycle.md

@@ -151,22 +151,15 @@ attention with both the original and repair artifacts recorded.
 
 ## Dashboard
 
-The dashboard is monitor-only:
+Dash is the human-first local control surface:
 
 ```sh
-night-shift start --ui
-night-shift resume --ui
+night-shift dash
 ```
 
-Night Shift binds to `127.0.0.1`, prefers port `8787`, and serves:
+Night Shift binds Dash to `127.0.0.1` for the current repository and serves:
 
-- run history for the current repository
-- run summary metadata
-- repo-state summary for review-driven runs, including open PR counts and drift
-- confidence posture and provenance path
-- task status
-- event timeline
-- report content
-
-There are no browser-side controls for starting or resuming runs in this
-version.
+- structured bootstrap state for initialization, runs, task DAG metadata, repo-state drift, confidence, runtime identities, report, and provenance references
+- SSE-first live updates for the selected repository state
+- browser command endpoints for `init`, `plan`, `plan --from-reviews`, `resolve`, `start`, and `resume`
+- raw artifact and audit routes for reports, provenance, logs, payloads, and runtime handoff files

src/night_shift/app.gleam

@@ -49,6 +49,11 @@ pub fn run(command: types.Command) -> Nil {
         <> "\n"
         <> crate_summary(types.default_config()),
       )
+    types.Dash ->
+      case dashboard_session.start(repo_root) {
+        Ok(Nil) -> Nil
+        Error(message) -> io.println(message)
+      }
     types.Demo(ui) ->
       case demo.run(ui) {
         Ok(summary) -> io.println(summary)
@@ -130,7 +135,6 @@ fn run_initialized_command(
         Error(message) -> message
       })
     types.Start(run, False) -> io.println(start(repo_root, run, config))
-    types.Start(run, True) -> start_with_ui(repo_root, run, config)
     types.Status(run) -> io.println(status(repo_root, run, config))
     types.Report(run) -> io.println(report(repo_root, run, config))
     types.Provenance(run, task_id, format) ->
@@ -139,8 +143,11 @@ fn run_initialized_command(
     types.Resolve(run) -> io.println(resolve(repo_root, run, config))
     types.Resume(run, False, False) ->
       io.println(resume(repo_root, run, config))
-    types.Resume(run, True, False) -> resume_with_ui(repo_root, run, config)
     types.Resume(run, False, True) -> io.println(doctor(repo_root, run, config))
+    types.Start(_, True) | types.Resume(_, True, False) ->
+      io.println(
+        "The `--ui` entrypoint was replaced by `night-shift dash`.",
+      )
     _ -> io.println("Unsupported command.")
   }
 }
@@ -330,28 +337,6 @@ fn reset(repo_root: String, assume_yes: Bool, force: Bool) -> String {
   }
 }
 
-fn start_with_ui(
-  repo_root: String,
-  selector: types.RunSelector,
-  config: types.Config,
-) -> Nil {
-  case dashboard_session.start(repo_root, selector, config) {
-    Ok(Nil) -> Nil
-    Error(message) -> io.println(message)
-  }
-}
-
-fn resume_with_ui(
-  repo_root: String,
-  run: types.RunSelector,
-  config: types.Config,
-) -> Nil {
-  case dashboard_session.resume(repo_root, run, config) {
-    Ok(Nil) -> Nil
-    Error(message) -> io.println(message)
-  }
-}
-
 fn review_profile_deprecation_fragment(review_profile: String) -> String {
   case review_profile {
     "" -> ""

src/night_shift/cli.gleam

@@ -10,18 +10,19 @@ pub fn usage() -> String {
   <> "\n"
   <> "Commands:\n"
   <> "  --demo [--ui]\n"
+  <> "  dash\n"
   <> "  init [--profile <name>] [--provider <codex|cursor>] [--model <id>] [--reasoning <low|medium|high|xhigh>] [--yes] [--generate-setup]\n"
   <> "    Prompts interactively for provider, model, and initial worktree setup when those answers are not supplied.\n"
   <> "  reset [--yes] [--force]\n"
   <> "  plan --notes <file-or-inline-text> [--doc <path>] [--profile <name>] [--provider <codex|cursor>] [--model <id>] [--reasoning <low|medium|high|xhigh>]\n"
   <> "  plan --from-reviews [--notes <file-or-inline-text>] [--doc <path>] [--profile <name>] [--provider <codex|cursor>] [--model <id>] [--reasoning <low|medium|high|xhigh>]\n"
-  <> "  start [--run <id>|latest] [--ui]\n"
+  <> "  start [--run <id>|latest]\n"
   <> "  status [--run <id>|latest]\n"
   <> "  report [--run <id>|latest]\n"
   <> "  provenance [--run <id>|latest] [--task <task-id>] [--format <json|md>]\n"
   <> "  doctor [--run <id>|latest]\n"
   <> "  resolve [--run <id>|latest]\n"
-  <> "  resume [--run <id>|latest] [--ui|--explain]\n"
+  <> "  resume [--run <id>|latest] [--explain]\n"
 }
 
 /// Parse raw command-line arguments into a `Command`.
@@ -44,6 +45,7 @@ pub fn parse(args: List(String)) -> Result(types.Command, String) {
       case args {
         [] -> Ok(types.Help)
         ["help", ..] -> Ok(types.Help)
+        ["dash"] -> Ok(types.Dash)
         ["init", ..rest] -> parse_init(rest)
         ["reset", ..rest] -> parse_reset(rest)
         ["plan", ..rest] -> parse_plan(rest)
@@ -254,7 +256,6 @@ fn parse_start_flags(
       parse_start_flags(rest, types.LatestRun, ui_enabled)
     ["--run", run_id, ..rest] ->
       parse_start_flags(rest, types.RunId(run_id), ui_enabled)
-    ["--ui", ..rest] -> parse_start_flags(rest, run, True)
     [flag, ..] -> Error("Unsupported start flag: " <> flag)
   }
 }
@@ -270,16 +271,11 @@ fn parse_resume_flags(
   explain_only: Bool,
 ) -> Result(types.Command, String) {
   case args {
-    [] ->
-      case ui_enabled && explain_only {
-        True -> Error("`resume --explain` cannot be combined with `--ui`.")
-        False -> Ok(types.Resume(run, ui_enabled, explain_only))
-      }
+    [] -> Ok(types.Resume(run, ui_enabled, explain_only))
     ["--run", "latest", ..rest] ->
       parse_resume_flags(rest, types.LatestRun, ui_enabled, explain_only)
     ["--run", run_id, ..rest] ->
       parse_resume_flags(rest, types.RunId(run_id), ui_enabled, explain_only)
-    ["--ui", ..rest] -> parse_resume_flags(rest, run, True, explain_only)
     ["--explain", ..rest] -> parse_resume_flags(rest, run, ui_enabled, True)
     [flag, ..] -> Error("Unsupported flag: " <> flag)
   }