pantheon-opencode 1.3.3 → 1.3.5-beta.52.c75b04d

package/README.md

@@ -2,7 +2,7 @@
 <h1 align="center">Pantheon</h1>
 
 <p align="center">
-  <a href="CHANGELOG.md"><img src="https://img.shields.io/badge/version-v1.0.0-blue" alt="Version"></a>
+  <a href="CHANGELOG.md"><img src="https://img.shields.io/badge/version-v1.3.4-blue" alt="Version"></a>
   <a href="LICENSE"><img src="https://img.shields.io/badge/license-MIT-blue" alt="License"></a>
   <a href="src/agents/README.md"><img src="https://img.shields.io/badge/agents-14-purple" alt="Agents"></a>
   <a href="src/skills/README.md"><img src="https://img.shields.io/badge/skills-21-orange" alt="Skills"></a>
@@ -183,6 +183,39 @@ export OPENCODE_EXPERIMENTAL_BACKGROUND_SUBAGENTS=true
 opencode
 ```
 
+### OpenCode V1/V2 — Dual Version
+
+Pantheon's shared config is **V1-shaped and valid under both OpenCode versions**:
+
+| | V1 | V2 (beta) |
+|---|---|---|
+| Binary | `opencode` (1.18.18) | `opencode2` (v0.0.0-next-17444) |
+| Config locations | `~/.config/opencode/`, `<project>/opencode.json`, `.opencode/` | Same — V2 reads the same locations |
+| Config handling | Loads V1 shape directly | Normalizes V1 fields **in memory** — no rewrite, no native V2 conversion |
+| State DB | `~/.local/share/opencode/opencode.db` | `~/.local/share/opencode/opencode-v2.db` (isolated via `OPENCODE_DB`) |
+| Service port | 49374 | 49375 (must differ from V1) |
+
+The `pantheon_cost` report preserves the V1 database default. Set
+`PANTHEON_OPENCODE_VERSION=v1` or `v2` to select the corresponding state DB
+explicitly (`opencode.db` or `opencode-v2.db`). For custom installations,
+`PANTHEON_COST_DB=/absolute/path/to/opencode.db` takes precedence over the
+version selector; an explicit `dbPath` supplied by the tool caller takes
+precedence over both. The resolver never probes the other version's database,
+and reports an actionable error when the selected DB is missing or has an
+incompatible schema.
+
+**What changed for V2 compatibility (Phase 3):**
+
+- `subagent_depth` moved from top-level to `experimental.subagent_depth` — V2 silently ignores the top-level key. The installer migrates existing configs automatically.
+- Instruction content consolidated into `AGENTS.md` (generated by `scripts/build-agents-md.mjs` from `src/instructions/*.instructions.md`) — V2 accepts but does **not** load the `instructions` config key; both versions auto-load `AGENTS.md`.
+- `npx pantheon-opencode init --version v2` labels an install for V2 (informational — the config is shared).
+
+**Known V2 beta gaps:**
+
+- `instructions` config key accepted but not loaded → content lives in `AGENTS.md`.
+- `formatter` / `lsp` config not executed.
+- `--subtask` delegation flag is a no-op.
+
 ### Background Delegation
 
 Zeus (and other root sessions) can dispatch agents as **background child sessions**
@@ -196,15 +229,27 @@ via three delegation tools, tracked on a persistent job board
 | `pantheon_delegation_read` | `{id}` | Blocks (`waitForTerminal`, up to 15 min) until the delegation reaches a terminal state, returns the report markdown, and marks the job reconciled. Resolves by alias or task ID. |
 | `pantheon_delegation_list` | `{}` | Lists the current session's delegations with `[unread]` on finished, unreconciled jobs. |
 
-**Notification model — no polling, no client push.** When a job reaches a
-terminal state (completion observed via `session.idle`/`session.error` on the
-child), a self-contained `<task-notification>` block is queued in memory for
-the parent session and injected into the parent's **next** `chat.message` hook
-fire (prepended onto the first text part — the graceful-degradation channel the
-TUI toast workaround uses). If the parent never sends another message, the
-notification stays queued. The `<task-notification>` carries the task ID, alias,
-agent, state, description and a `Result:`-prefixed summary so the parent can act
-without an extra tool call.
+**Agent activity visibility.** While `pantheon_delegation_read` blocks on a
+running job, it samples the child session's messages every ~2s and appends the
+collected lines to the report as a trailing `## Agent Activity` section (latest
+tool calls with truncated args, or user/assistant text, capped at ~200 chars per
+line) — so you can see what the agent is doing during the wait. Likewise,
+`pantheon_delegation_list` shows a `last activity:` line for running jobs.
+Fail-open: if the child session messages are unavailable (or empty), the read
+returns the report exactly as before — the activity sampling never breaks the
+delegation read.
+
+**Completion visibility — no chat injection.** User policy: ZERO delegation
+notifications in the chat transcript — no `<task-notification>` block is ever
+injected into a `chat.message` output (the old queue + flush channel was
+removed). When a job reaches a terminal state (completion observed via
+`session.idle`/`session.error` on the child, or the timeout finalize path), the
+plugin writes a file-only audit log line (echo opt-in via `PANTHEON_HOOKS_LOG`).
+Completion visibility lives in the legitimate channels: the board `[unread]`
+marker (`pantheon_delegation_list`), `pantheon_delegation_read`, TUI toasts
+(pantheon-hooks, `PANTHEON_TOASTS` gate) and compaction carry-forward. Reconcile
+is an acknowledgment, not a completion — it never re-fires the terminal
+transition.
 
 **Timeout:** `background_delegation.timeout_ms` (default `900000` = 15 min). A job
 that has not reached a terminal state is finalized as `error`/`timedOut`.
@@ -219,13 +264,80 @@ terminal delegations (capped at `background_delegation.max_compaction_items`,
 default 10) are carried into the compacted context so the parent doesn't lose
 track of in-flight work.
 
+**Compaction summary v2 (1.3.4):** the `experimental.session.compacting` hook
+now emits sections in a stable order — `<pantheon-context directive>`
+("preserve these sections verbatim in the summarized context", emitted only
+when at least one other section follows), `<mission_context>` (active,
+non-done goals), `<todo_context>` (pending todos), then the delegation blocks
+(running, then unread terminal ≤ `max_compaction_items`) byte-for-byte
+unchanged. Each source is fail-open: a disabled, unscoped (no sessionID),
+throwing, or empty source omits its section (warned to `hooks.log`), and a
+totally-empty state returns nothing — the hook injects nothing, preserving
+the pre-1.3.4 behavior.
+
+**Post-compaction todo preservation (1.3.4):** the session's todo list is
+captured via the `session.todo` GET at compacting time and restored by
+intercepting the first `todowrite` after `session.compacted` — opencode
+1.18.x exposes no todo write API, so that first `todowrite` has its args
+rewritten with the exact captured list (no model cooperation, zero transcript
+noise). Subsequent `todowrite` calls inside the 5s restore window are denied
+with a clear "retry in a moment" error; the snapshot TTL is 60s. Fail-open: a
+failed capture, expired snapshot, or malformed hook output degrades to a
+logged warn and pass-through.
+
+**Post-compaction state re-assertion (1.3.4):** on `session.compacted`, fresh
+state — running/unread delegations from the board + active goals, capped at 10
+lines — is re-injected as a `<system-reminder>` into the session's next
+message via the shared chat-reminder buffer (`chat-reminders.ts`, the same P0
+messageID guard that protects subagent fires). A session with nothing to
+assert is a silent skip.
+
+**Preemptive compaction check (1.3.4):** a pure threshold core
+(`preemptive-compact.ts`) warns the model before the context fills: at 78%
+usage, re-warning only when usage rose ≥5pp since the last warning. It is
+dormant / ready-to-wire — opencode 1.18.x exposes no runtime context-usage
+percentage, so nothing observes it yet; when a source appears, the caller
+wires it in (the enqueue callback is injected).
+
+**Model API-key validation (1.3.4):** `pantheon_delegate` gates the resolved
+child model's provider before dispatching (single source of truth:
+`routing.yml` preset definitions' `apiKeyEnv` — the same check `applyPreset`
+enforces at startup). If an **auto-resolved** model (`options.agentModels` or
+the active preset) points to a provider that requires an API key
+(`apiKeyEnv`) and the env var is unset, the delegate falls back to
+`opencode/deepseek-v4-flash-free` (validated the same way). If the fallback is
+also unusable, the tool returns a clear error **text** (never throws) naming
+the missing env var — and registers **no job** on the board. An **explicit**
+`model` passed by the caller is always respected (warned, not overridden).
+Setting `PANTHEON_MODEL_PRESET` to a preset whose providers have keys, or
+filling the required `PANTHEON_*_API_KEY` env var for the preset's provider,
+resolves the fallback path. If nothing resolves at all (no model, no preset),
+the child keeps using opencode's default model (warned).
+
+**agentModels wiring (1.3.4):** the delegate now also resolves the child model
+via `routing.yml` — `loadRoutingAgentModels` extracts the per-agent models of
+the FIRST-listed preset (the static default, `go-deepseek` today) and passes
+them as `options.agentModels`, branch (b) of the resolve precedence: explicit
+`model` > `options.agentModels` > active preset > opencode default. Delegation
+no longer depends exclusively on the active preset — a delegated child gets a
+sane per-agent model even when no preset is active. Fail-open: a missing or
+unparseable routing.yml yields `{}` (previous behavior, warned).
+
+**Delegation log hygiene (1.3.4):** `delegations.log` now records the real
+`task_id` (omitted when empty, never `""`) and a **numeric** `duration_ms`
+(null when unset) so downstream aggregation works. The idle-flush log entry is
+deduplicated: the flush logs a summary (count + aggregated line count) and the
+reminder content is logged exactly once, at `chat.message` delivery — no more
+duplicate lines for the same notification.
+
 **Env vars & config:**
 
 - `OPENCODE_EXPERIMENTAL_BACKGROUND_SUBAGENTS=true` — required (see launch block above).
 - `PANTHEON_TOASTS` — TUI toast gate for the delegation lifecycle signals.
   Default `{errors, delegations, council}`; `PANTHEON_TOASTS=off` disables all
-  TUI toasts (delegation signals then only surface via the chat.message
-  `<task-notification>`/`<system-reminder>` channel).
+  TUI toasts (delegation signals then only surface via the board `[unread]`
+  marker, `pantheon_delegation_read` and the on-disk audit log — never via
+  chat.message injection from the plugin).
 - `background_delegation` section in `src/routing.yml` — `timeout_ms`,
   `poll_ms`, `max_compaction_items`, `prune_ttl_ms`, `read_only_agents`,
   `session_max`, `retry_count`.
@@ -233,6 +345,30 @@ track of in-flight work.
 See `src/agents/zeus.md` for the agent-facing delegation protocol and
 `src/pantheon/delegation.ts` for the tool implementations.
 
+### Real-time Delegations panel (1.3.4)
+
+The TUI sidebar ships a live **Delegations** panel (open by default, above the
+collapsed Sessions list). Its primary source is `api.client.session.children`,
+so it shows **both** kinds of background work in one place:
+
+- **`pantheon_delegate` children** — rendered with their board alias tag
+  (`[apo-1]`) and enriched from the `.pantheon/delegations/` report.
+- **Native `task()` children** — rendered with a distinct `[task]` tag (info
+  color) when no board report exists for the session.
+
+Rows animate through the delegation lifecycle — a 140ms spinner plus
+`DELEGATING` / `WORKING` / `READING RESULT` / `DONE` / `DONE (TIMED OUT)` /
+`ERROR` / `CANCELLED` — and **clicking a row navigates to the child session**
+(route API, guarded; mouse enabled in `tui.json`). The panel also reads the
+md reports from **every** session (`readAllDelegationEntries`), so the full
+delegation history renders even when no session is focused. Refresh is driven
+by session events + `message.part.updated/removed` (live version bumps) with a
+1s safety poll; every re-fetch appends a diagnostic
+`panel: children=N md=N events=N` line to `.pantheon/logs/hooks.log`
+(silence-by-default policy — see `PANTHEON_HOOKS_LOG`). Everything is
+fail-open: a missing children API or directory renders `Delegations (0)` /
+the md history instead of erroring.
+
 > **Modes:** Interactive TUI with checkbox selection (default TTY), or `--headless` for scripts/CI.
 > **Minimal:** `--headless --no-mcp` installs only agents (~2s).
 > **Full:** `--headless` also creates Python venv + MCP servers (memory, persistence, KV, vision).
@@ -750,9 +886,15 @@ to Mnemosyne so counts stay accurate and consistent.
 
 ### What CI enforces automatically
 
-`release.yml` validates version consistency across all manifests (`package.json`,
-`plugin.json`, `CHANGELOG.md`, and the README badge) before publishing. If they diverge,
-the release is blocked until Mnemosyne reconciles them.
+`release.yml` validates version consistency across the package manifests
+(`package.json`, `plugin.json`, `pyproject.toml`, and the TUI package) before
+publishing. If they diverge, the release is blocked until they are reconciled.
+
+Beta releases are calculated from npm's published `latest` at workflow runtime:
+the default is the next patch, formatted as
+`<next-stable>-beta.<PR>.<short-sha>`. Labels `release:beta:minor` and
+`release:beta:major` request minor or major calculation. The workflow owns the
+manifest update, tag, and publish; it does not assume a fixed future version.
 
 ### Anti-patterns
 

package/bin/pantheon-init.mjs

@@ -6,205 +6,262 @@
  *
  * Thin CLI wrapper that parses arguments and delegates to
  * installOpenCode() from scripts/install/opencode.mjs.
+ *
+ * Dual-version install (Phase 3): the generated config is V1-shaped and valid
+ * under BOTH OpenCode V1 (`opencode` 1.18.x) and V2 (`opencode2`
+ * v0.0.0-next-17444). V2 reads the same config locations and normalizes V1
+ * fields in memory — no native V2 conversion. `--version v2` labels the
+ * install for the operator; state isolation is a runtime concern
+ * (OPENCODE_DB ~/.local/share/opencode/opencode-v2.db, service port 49375).
  */
-import fs from 'fs';
-import os from 'os';
-import path from 'path';
-import { fileURLToPath } from 'url';
+import fs from 'fs'
+import os from 'os'
+import path from 'path'
+import { fileURLToPath } from 'url'
 
-const __dirname = path.dirname(fileURLToPath(import.meta.url));
-const ROOT = path.resolve(__dirname, '..');
+const __dirname = path.dirname(fileURLToPath(import.meta.url))
+const ROOT = path.resolve(__dirname, '..')
 
-const REQUIRED_NODE_MAJOR = 18;
+const REQUIRED_NODE_MAJOR = 18
 if (parseInt(process.versions.node.split('.')[0], 10) < REQUIRED_NODE_MAJOR) {
-  console.error(`❌ Node.js >= ${REQUIRED_NODE_MAJOR} required (current: ${process.versions.node})`);
-  process.exit(1);
+  console.error(`❌ Node.js >= ${REQUIRED_NODE_MAJOR} required (current: ${process.versions.node})`)
+  process.exit(1)
+}
+
+function readVersion() {
+  try {
+    const pkgPath = path.join(ROOT, 'package.json')
+    return JSON.parse(fs.readFileSync(pkgPath, 'utf8')).version || '?'
+  } catch {
+    return '?'
+  }
 }
 
 function printUsage() {
-  console.log('Pantheon OpenCode — Multi-agent orchestration platform');
-  console.log('');
-  console.log('Usage:');
-  console.log('  npx pantheon-opencode init              # Install globally');
-  console.log('  npx pantheon-opencode init --project    # Install in project');
-  console.log('  npx pantheon-opencode init --dry-run    # Preview only');
-  console.log('  npx pantheon-opencode init --no-mcp       # Skip MCP + venv');
-  console.log('  npx pantheon-opencode init --force        # Overwrite + recreate venv');
-  console.log('  npx pantheon-opencode init --doctor       # Run health check after install');
-  console.log('  npx pantheon-opencode init --interactive  # Force interactive TUI mode');
-  console.log('  npx pantheon-opencode init --headless     # Force non-interactive mode');
-  console.log('  npx pantheon-opencode init -y             # Skip confirmations, use defaults');
-  console.log('  npx pantheon-opencode init --preset <name> # Install and activate model preset');
-  console.log('  npx pantheon-opencode set-tier <name>      # Set active model preset (global)');
-  console.log('  npx pantheon-opencode set-tier <name> --project  # Set active model preset (project)');
-  console.log('  npx pantheon-opencode prune               # List legacy artifacts (dry-run)');
-  console.log('  npx pantheon-opencode prune --apply       # Remove stale backups');
-  console.log('  npx pantheon-opencode --help              # Show this help');
+  console.log('Pantheon OpenCode — Multi-agent orchestration platform')
+  console.log('')
+  console.log('Usage:')
+  console.log('  npx pantheon-opencode init              # Install globally')
+  console.log('  npx pantheon-opencode init --project    # Install in project')
+  console.log('  npx pantheon-opencode init --dry-run    # Preview only')
+  console.log('  npx pantheon-opencode init --no-mcp       # Skip MCP + venv')
+  console.log('  npx pantheon-opencode init --force        # Overwrite + recreate venv')
+  console.log('  npx pantheon-opencode init --doctor       # Run health check after install')
+  console.log('  npx pantheon-opencode init --interactive  # Force interactive TUI mode')
+  console.log('  npx pantheon-opencode init --headless     # Force non-interactive mode')
+  console.log('  npx pantheon-opencode init -y             # Skip confirmations, use defaults')
+  console.log('  npx pantheon-opencode init --version v2    # Label install for OpenCode V2 (beta)')
+  console.log('  npx pantheon-opencode init --preset <name> # Install and activate model preset')
+  console.log('  npx pantheon-opencode set-tier <name>      # Set active model preset (global)')
+  console.log(
+    '  npx pantheon-opencode set-tier <name> --project  # Set active model preset (project)',
+  )
+  console.log('  npx pantheon-opencode prune               # List legacy artifacts (dry-run)')
+  console.log('  npx pantheon-opencode prune --apply       # Remove stale backups')
+  console.log('  npx pantheon-opencode doctor [--profile]  # Run health check')
+  console.log('  npx pantheon-opencode --version, -v       # Print version and exit')
+  console.log('  npx pantheon-opencode --help              # Show this help')
 }
 
 async function main() {
-  const args = process.argv.slice(2);
-  const command = args[0];
+  const args = process.argv.slice(2)
+  const command = args[0]
+
+  // --version / -v: print the package version and exit. Must run before any
+  // install/help dispatch so it never touches the filesystem beyond reading
+  // package.json.
+  if (command === '--version' || command === '-v') {
+    console.log(readVersion())
+    process.exit(0)
+  }
 
   // prune handles its own --help (delegated to scripts/prune.mjs), so its
   // dispatch must run BEFORE the generic --help check.
   if (command === 'prune') {
-    const { spawnSync } = await import('child_process');
-    const pruneScript = path.join(ROOT, 'scripts', 'prune.mjs');
+    const { spawnSync } = await import('child_process')
+    const pruneScript = path.join(ROOT, 'scripts', 'prune.mjs')
     const result = spawnSync(process.execPath, [pruneScript, ...args.slice(1)], {
       stdio: 'inherit',
       cwd: ROOT,
-    });
-    process.exit(result.status ?? 1);
+    })
+    process.exit(result.status ?? 1)
   }
 
-  if (args.includes('--help')) { printUsage(); process.exit(0); }
+  // doctor delegates to scripts/doctor.mjs (same spawnSync pattern as prune).
+  // Forwarding the FULL args (profile, target, --fix, ...) keeps the packaged
+  // CLI a thin wrapper; the doctor banner reports the resolved profile.
+  if (command === 'doctor') {
+    const { spawnSync } = await import('child_process')
+    const doctorScript = path.join(ROOT, 'scripts', 'doctor.mjs')
+    const result = spawnSync(process.execPath, [doctorScript, ...args.slice(1)], {
+      stdio: 'inherit',
+      cwd: ROOT,
+    })
+    process.exit(result.status ?? 1)
+  }
+
+  if (args.includes('--help')) {
+    printUsage()
+    process.exit(0)
+  }
 
   if (command === 'set-tier') {
-    const isProject = args.includes('--project');
-    const isDryRun = args.includes('--dry-run');
-    const name = args[1];
+    const isProject = args.includes('--project')
+    const isDryRun = args.includes('--dry-run')
+    const name = args[1]
 
-    const { loadPresetDefs } = await import('../src/pantheon/presets.mjs');
-    const { writeActivePreset } = await import('../scripts/install/model-picker.mjs');
-    const presets = loadPresetDefs();
-    const presetList = `none, ${Object.keys(presets).join(', ')}`;
+    const { loadPresetDefs } = await import('../src/pantheon/presets.mjs')
+    const { writeActivePreset } = await import('../scripts/install/model-picker.mjs')
+    const presets = loadPresetDefs()
+    const presetList = `none, ${Object.keys(presets).join(', ')}`
 
     if (!name) {
-      console.error('❌ set-tier requires a preset name');
-      console.error(`   Available: ${presetList}`);
-      process.exit(1);
+      console.error('❌ set-tier requires a preset name')
+      console.error(`   Available: ${presetList}`)
+      process.exit(1)
     }
     if (name !== 'none' && !presets[name]) {
-      console.error(`❌ Unknown preset "${name}"`);
-      console.error(`   Available: ${presetList}`);
-      process.exit(1);
+      console.error(`❌ Unknown preset "${name}"`)
+      console.error(`   Available: ${presetList}`)
+      process.exit(1)
     }
 
     // Fail-fast: every provider's API key env var must be set before writing
     if (name !== 'none') {
       for (const [pid, provider] of Object.entries(presets[name].providers || {})) {
         if (!process.env[provider.apiKeyEnv]) {
-          console.error(`❌ Missing required API key configuration for provider "${pid}"`);
-          console.error('   Set the provider API key environment variable and retry.');
-          process.exit(1);
+          console.error(`❌ Missing required API key configuration for provider "${pid}"`)
+          console.error('   Set the provider API key environment variable and retry.')
+          process.exit(1)
         }
       }
     }
 
     const presetDir = isProject
       ? process.cwd()
-      : path.join(process.env.XDG_CONFIG_HOME || path.join(os.homedir(), '.config'), 'opencode');
-    const filePath = path.join(presetDir, '.pantheon', 'active-preset.json');
-    const bakPath = path.join(presetDir, '.pantheon', 'active-preset.json.bak');
+      : path.join(process.env.XDG_CONFIG_HOME || path.join(os.homedir(), '.config'), 'opencode')
+    const filePath = path.join(presetDir, '.pantheon', 'active-preset.json')
+    const bakPath = path.join(presetDir, '.pantheon', 'active-preset.json.bak')
 
-    let prevName = '(none)';
+    let prevName = '(none)'
     if (fs.existsSync(filePath)) {
       try {
-        prevName = JSON.parse(fs.readFileSync(filePath, 'utf8')).preset || '(unknown)';
-      } catch { prevName = '(unknown)'; }
+        prevName = JSON.parse(fs.readFileSync(filePath, 'utf8')).preset || '(unknown)'
+      } catch {
+        prevName = '(unknown)'
+      }
     }
 
     if (name === 'none') {
       if (isDryRun) {
-        console.log('[dry-run] Would clear model preset');
+        console.log('[dry-run] Would clear model preset')
       } else {
         if (fs.existsSync(filePath)) {
-          fs.mkdirSync(path.dirname(bakPath), { recursive: true });
-          fs.copyFileSync(filePath, bakPath);
-          fs.rmSync(filePath, { force: true });
+          fs.mkdirSync(path.dirname(bakPath), { recursive: true })
+          fs.copyFileSync(filePath, bakPath)
+          fs.rmSync(filePath, { force: true })
         }
-        console.log('Model preset cleared. Using opencode defaults.');
-        console.log(`   Backup: ${bakPath}`);
+        console.log('Model preset cleared. Using opencode defaults.')
+        console.log(`   Backup: ${bakPath}`)
       }
-      return;
+      return
     }
 
-    const result = writeActivePreset(presetDir, name, { dryRun: isDryRun, source: 'cli' });
+    const result = writeActivePreset(presetDir, name, { dryRun: isDryRun, source: 'cli' })
 
-    console.log(`Model preset set: ${prevName} → ${name}`);
+    console.log(`Model preset set: ${prevName} → ${name}`)
     for (const [agent, spec] of Object.entries(presets[name].agents || {})) {
-      console.log(`   ${agent}: ${spec.model} [${spec.reasoning_effort}]`);
+      console.log(`   ${agent}: ${spec.model} [${spec.reasoning_effort}]`)
     }
-    const providerNames = Object.keys(presets[name].providers || {});
-    console.log(`   Providers: ${providerNames.join(', ') || 'none'}`);
-    if (result.backupPath) console.log(`   Backup: ${result.backupPath}`);
-    return;
+    const providerNames = Object.keys(presets[name].providers || {})
+    console.log(`   Providers: ${providerNames.join(', ') || 'none'}`)
+    if (result.backupPath) console.log(`   Backup: ${result.backupPath}`)
+    return
   }
 
   if (command === 'init' || !command) {
-    const isProject = args.includes('--project');
-    const isDryRun = args.includes('--dry-run');
-    const skipMCP = args.includes('--no-mcp');
-    const forceReinstall = args.includes('--force');
-    const runDoctor = args.includes('--doctor');
-    const forceInteractive = args.includes('--interactive');
-    const forceHeadless = args.includes('--headless');
-    const autoYes = args.includes('--yes') || args.includes('-y');
-    const presetIndex = args.indexOf('--preset');
-    const presetOpt = presetIndex >= 0 ? (args[presetIndex + 1] ?? null) : null;
-
-    const components = ['agents', 'skills', 'instructions', 'commands', 'plugins'];
-    if (!skipMCP) components.push('runtime');
+    const isProject = args.includes('--project')
+    const isDryRun = args.includes('--dry-run')
+    const skipMCP = args.includes('--no-mcp')
+    const forceReinstall = args.includes('--force')
+    const runDoctor = args.includes('--doctor')
+    const forceInteractive = args.includes('--interactive')
+    const forceHeadless = args.includes('--headless')
+    const autoYes = args.includes('--yes') || args.includes('-y')
+    const presetIndex = args.indexOf('--preset')
+    const presetOpt = presetIndex >= 0 ? (args[presetIndex + 1] ?? null) : null
+    // --version v1|v2 labels the install target (informational; the config is
+    // shared and V1-shaped under both versions). Invalid values fail fast.
+    const versionIndex = args.indexOf('--version')
+    const versionOpt = versionIndex >= 0 ? (args[versionIndex + 1] ?? null) : null
+    if (versionOpt !== null && versionOpt !== 'v1' && versionOpt !== 'v2') {
+      console.error(`❌ Invalid --version "${versionOpt}" — expected v1 or v2`)
+      process.exit(1)
+    }
+
+    const components = ['agents', 'skills', 'instructions', 'commands', 'plugins']
+    if (!skipMCP) components.push('runtime')
 
     // Version info
-    let version = '?';
-    try {
-      const pkgPath = path.join(ROOT, 'package.json');
-      version = JSON.parse(fs.readFileSync(pkgPath, 'utf8')).version || '?';
-    } catch { /* use default */ }
+    const version = readVersion()
 
-    console.log('');
+    console.log('')
     if (!forceInteractive || isDryRun) {
-      console.log(`Pantheon OpenCode v${version} — ${isDryRun ? 'DRY RUN' : 'Installing...'}`);
-      console.log('');
+      console.log(`Pantheon OpenCode v${version} — ${isDryRun ? 'DRY RUN' : 'Installing...'}`)
+      console.log('')
     }
 
     try {
-      const { installOpenCode } = await import('../scripts/install/opencode.mjs');
-      const target = isProject ? process.cwd() : undefined;
+      const { installOpenCode } = await import('../scripts/install/opencode.mjs')
+      const target = isProject ? process.cwd() : undefined
       await installOpenCode(target, isDryRun, forceReinstall, components, {
         interactive: forceInteractive,
         headless: forceHeadless,
         yes: autoYes,
         preset: presetOpt,
-      });
+        version: versionOpt ?? 'v1',
+      })
     } catch (err) {
-      console.error(`❌ Installation failed: ${err.message}\n   Run with --no-mcp to skip Python dependencies:\n     npx pantheon-opencode init --no-mcp\n   Or retry with --force to recreate the venv:\n     npx pantheon-opencode init --force`);
-      process.exit(1);
+      console.error(
+        `❌ Installation failed: ${err.message}\n   Run with --no-mcp to skip Python dependencies:\n     npx pantheon-opencode init --no-mcp\n   Or retry with --force to recreate the venv:\n     npx pantheon-opencode init --force`,
+      )
+      process.exit(1)
     }
 
     if (runDoctor && !isDryRun) {
-      console.log('');
-      console.log('  Running health check...');
+      console.log('')
+      console.log('  Running health check...')
       try {
-        const { spawnSync } = await import('child_process');
-        const doctorScript = path.join(ROOT, 'scripts', 'doctor.mjs');
-        spawnSync(process.execPath, [doctorScript], { stdio: 'inherit', cwd: ROOT });
+        const { spawnSync } = await import('child_process')
+        const doctorScript = path.join(ROOT, 'scripts', 'doctor.mjs')
+        spawnSync(process.execPath, [doctorScript], { stdio: 'inherit', cwd: ROOT })
       } catch {
-        console.log('  ⚠️  Could not run doctor.mjs');
+        console.log('  ⚠️  Could not run doctor.mjs')
       }
     }
 
-    console.log('');
-    console.log('━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━');
-    console.log(`  ✅ Pantheon OpenCode v${version} installed!`);
-    console.log('━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━');
-    console.log('');
-    console.log('  Next steps:');
-    console.log('  1. Verify installation:');
-    console.log('     npx pantheon-opencode doctor');
-    console.log('  2. Launch your AI coding tool (OpenCode, Claude Code, Cursor, etc.)');
-    console.log('  3. Invoke agents with @agent-name in chat');
-    console.log('  4. For project-local install:');
-    console.log('     npx pantheon-opencode init --project');
-    console.log('');
-
-    return;
+    console.log('')
+    console.log('━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━')
+    console.log(`  ✅ Pantheon OpenCode v${version} installed!`)
+    console.log('━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━')
+    console.log('')
+    console.log('  Next steps:')
+    console.log('  1. Verify installation:')
+    console.log('     npx pantheon-opencode doctor')
+    console.log('  2. Launch your AI coding tool (OpenCode, Claude Code, Cursor, etc.)')
+    console.log('  3. Invoke agents with @agent-name in chat')
+    console.log('  4. For project-local install:')
+    console.log('     npx pantheon-opencode init --project')
+    console.log('')
+
+    return
   }
 
-  printUsage();
-  process.exit(1);
+  printUsage()
+  process.exit(1)
 }
 
-main().catch(err => { console.error('Error:', err.stack); process.exit(1); });
+main().catch((err) => {
+  console.error('Error:', err.stack)
+  process.exit(1)
+})

package/opencode.json

@@ -11,7 +11,9 @@
     }
   },
   "default_agent": "zeus",
-  "subagent_depth": 2,
+  "experimental": {
+    "subagent_depth": 2
+  },
   "model": "opencode-go/deepseek-v4-flash",
   "small_model": "opencode-go/deepseek-v4-flash",
   "permission": {
@@ -28,11 +30,6 @@
       "pip *": "allow"
     }
   },
-  "instructions": ["AGENTS.md", "src/instructions/*.instructions.md"],
-  "plugin": [
-    "/home/ils15/pantheon/src/plugin.ts",
-    "/home/ils15/pantheon/src/plugins/pantheon-hooks.ts"
-  ],
   "mcp": {
     "bifrost": {
       "type": "remote",