@hueyexe/opencode-ensemble 0.5.0 → 0.5.1

package/README.md

@@ -1,21 +1,21 @@
 # OpenCode Ensemble
 
-Run a team of parallel AI agents in OpenCode. Each agent gets its own session, context window, and task. They coordinate through messaging and a shared task board.
+Run parallel AI agents in OpenCode. Each agent gets its own session, context window, and task. They coordinate through messaging and a shared task board.
 
-Built as a plugin on the public OpenCode SDK. Zero internal dependencies.
+Plugin built on the public OpenCode SDK. No internal dependencies.
 
 ## What actually happens
 
-You ask the agent to do something complex. The agent decides it needs help, creates a team, and spawns teammates. Each teammate gets its own OpenCode session with a fresh context window and a specific task.
+You ask the agent to do something complex. It creates a team, spawns teammates, and they work in parallel. Each teammate runs in its own OpenCode session with a fresh context window.
 
-Here's a real interaction:
+A real interaction:
 
 ```
 You: "Add input validation to all API endpoints and write tests for each one."
 
 The lead agent:
 1. Creates a team called "validation"
-2. Adds tasks to the shared board: one per endpoint
+2. Adds tasks to the shared board, one per endpoint
 3. Spawns 3 teammates:
    - alice: validate user endpoints (POST /users, PUT /users/:id)
    - bob: validate order endpoints (POST /orders, PUT /orders/:id)
@@ -23,7 +23,7 @@ The lead agent:
 4. carol's tasks depend on alice and bob finishing first
 ```
 
-From here, things happen in parallel. Alice and bob work simultaneously in their own sessions. You see toast notifications as they make progress:
+Alice and bob work simultaneously. You see toast notifications as they progress:
 
 ```
 [toast] Teammate alice spawned (build)
@@ -31,18 +31,18 @@ From here, things happen in parallel. Alice and bob work simultaneously in their
 [toast] Teammate carol spawned (build)
 ```
 
-The teammates talk to each other and to the lead:
+Teammates talk to each other and to the lead:
 
 ```
 alice -> lead: "User validation done. Added zod schemas to POST /users and PUT /users/:id."
-bob -> lead: "Order validation done. Found an edge case in PUT /orders/:id — negative quantities were allowed."
+bob -> lead: "Order validation done. Found an edge case in PUT /orders/:id, negative quantities were allowed."
 bob -> alice: "Did you handle email format validation? I want to match the pattern for order contact emails."
-alice -> bob: "Yes, using z.string().email() — see src/validators/user.ts line 12."
+alice -> bob: "Yes, using z.string().email(). See src/validators/user.ts line 12."
 ```
 
-When alice and bob finish, carol's blocked tasks unblock automatically. Carol starts writing tests using the actual validation schemas alice and bob created.
+When alice and bob finish, carol's blocked tasks unblock automatically. Carol starts writing tests using the validation schemas they created.
 
-The lead keeps you updated throughout. You can check on things at any time:
+Check on things at any time:
 
 ```
 You: "How's the team doing?"
@@ -56,7 +56,7 @@ Lead calls team_status:
   Tasks: 5 total (3 completed, 1 in_progress, 1 pending)
 ```
 
-Want to see what carol is actually doing? The lead can switch your view to her session:
+Want to see what carol is doing? The lead can switch your view to her session:
 
 ```
 You: "Show me what carol is working on."
@@ -79,9 +79,9 @@ Lead: "All validation and tests are complete. 5 endpoints validated,
 
 ## Install
 
-Add the plugin to your OpenCode config. This can be project-level or global.
+Add the plugin to your OpenCode config. Project-level or global.
 
-**Project-level** — `opencode.json` in your project root:
+**Project-level** (`opencode.json` in your project root):
 
 ```json
 {
@@ -89,7 +89,7 @@ Add the plugin to your OpenCode config. This can be project-level or global.
 }
 ```
 
-**Global** — `~/.config/opencode/opencode.json`:
+**Global** (`~/.config/opencode/opencode.json`):
 
 ```json
 {
@@ -97,7 +97,7 @@ Add the plugin to your OpenCode config. This can be project-level or global.
 }
 ```
 
-That's it. OpenCode auto-installs npm plugins at startup (cached in `~/.cache/opencode/node_modules/`). No `npm install` or `bun add` needed.
+OpenCode auto-installs npm plugins at startup (cached in `~/.cache/opencode/node_modules/`). No `npm install` needed.
 
 ### Local development
 
@@ -115,15 +115,15 @@ export { default } from "/path/to/opencode-ensemble/src/index.ts"
 
 ## Tools
 
-13 tools total. The lead has all of them. Teammates get 6 (messaging + tasks).
+13 tools. The lead has all of them. Teammates get 6 (messaging + tasks).
 
 **Team lifecycle** (lead only)
 
 | Tool | What it does |
 |------|-------------|
 | `team_create` | Create a team. Caller becomes the lead. |
-| `team_spawn` | Start a new teammate with a task. |
-| `team_shutdown` | Ask a teammate to stop. |
+| `team_spawn` | Start a new teammate with a task. Supports `plan_approval` mode. |
+| `team_shutdown` | Ask a teammate to stop. Supports `force` flag for immediate abort. |
 | `team_cleanup` | Remove the team when done. |
 | `team_status` | See all members, their status, and a task summary. |
 | `team_view` | Switch the TUI to a teammate's session. |
@@ -132,7 +132,7 @@ export { default } from "/path/to/opencode-ensemble/src/index.ts"
 
 | Tool | What it does |
 |------|-------------|
-| `team_message` | Send a direct message to a teammate or the lead. |
+| `team_message` | Send a direct message to a teammate or the lead. Also handles plan approval/rejection. |
 | `team_broadcast` | Message everyone on the team. |
 | `team_results` | Retrieve full message content (messages to lead are truncated on delivery). |
 
@@ -147,28 +147,31 @@ export { default } from "/path/to/opencode-ensemble/src/index.ts"
 
 ## What you see in the TUI
 
-The plugin works within OpenCode's existing TUI. There's no custom team panel (that requires core TUI changes, which are [in progress upstream](https://github.com/sst/opencode)).
+The plugin works within OpenCode's existing TUI. No custom team panel (that requires core TUI changes, which are [in progress upstream](https://github.com/sst/opencode)).
 
-What you do get:
+What you get:
 
 - **Toast notifications** when teammates spawn, finish, error, shut down, or get rate-limited
 - **Working progress toasts** showing who's still active after every status change (e.g. "Working: alice, bob (2/3)")
 - **Rich tool titles** in the sidebar (e.g. "Spawned alice (build)", "Message -> bob", "Task board (3 tasks)")
 - **Session switching** via `team_view` to see any teammate's full chat log
-- **Status checks** via `team_status` for a snapshot of the whole team (also fires a toast)
+- **Status checks** via `team_status` for a snapshot of the whole team
 
-Teammate messages arrive in the lead's session as `[Team message from alice]: ...` blocks. They look like user messages in the TUI because that's how `promptAsync` delivery works. The content is clearly labeled with the sender's name.
+Teammate messages arrive in the lead's session as `[Team message from alice]: ...` blocks. They look like user messages because that's how `promptAsync` delivery works. Content is clearly labeled with the sender's name.
 
 ## Architecture
 
 - **SQLite** (`bun:sqlite`, WAL mode) for teams, members, tasks, and messages
 - **promptAsync** for message delivery: injects a message and starts the prompt loop in one call
-- **Git worktree isolation**: each teammate gets their own worktree by default, so multiple agents can edit files without conflicts. The lead merges branches after cleanup. Opt out with `worktree: false` for read-only agents.
+- **Git worktree isolation**: each teammate gets their own worktree by default, so multiple agents can edit files without conflicts. Opt out with `worktree: false` for read-only agents.
+- **System prompt injection**: the lead's system prompt includes team state (member statuses, task counts) on every LLM call. Teammates get a short role reminder.
+- **Compaction safety**: team context is preserved when OpenCode compacts long conversations
+- **Shell environment**: teammate shells get `ENSEMBLE_TEAM`, `ENSEMBLE_MEMBER`, `ENSEMBLE_ROLE`, and `ENSEMBLE_BRANCH` variables
 - **Sub-agent isolation**: teammates' sub-agents can't use team tools (parent chain tracking, max depth 10)
 - **Crash recovery**: stale busy members marked as errored on restart, orphaned sessions aborted, orphaned worktrees cleaned up, undelivered messages redelivered
 - **Spawn rollback**: if the initial prompt fails, the member, session, and worktree are all cleaned up
 - **Timeout watchdog**: teammates stuck busy beyond the TTL are automatically timed out and aborted
-- **Shutdown resilience**: abort failures handled gracefully, busy-after-shutdown triggers re-abort
+- **Graceful shutdown**: busy teammates receive a shutdown message and finish their current work. Use `force: true` to abort immediately.
 - **Rate limiting**: token bucket (configurable via `OPENCODE_ENSEMBLE_RATE_LIMIT`, default 10 tokens/sec)
 
 ## Configuration
@@ -187,34 +190,34 @@ OPENCODE_ENSEMBLE_TIMEOUT=3600000
 OPENCODE_ENSEMBLE_TIMEOUT=0
 ```
 
-### Best practices
+## Best practices
 
-- Start with 2-4 teammates for most tasks. More agents means more coordination overhead.
-- Give each teammate 3-5 specific, self-contained tasks. Vague prompts produce vague results.
-- Start with research before implementation — spawn an explore agent first to understand the codebase, then spawn build agents with that context.
-- Use `worktree: false` for read-only agents (research, review, code analysis). They don't need file isolation.
-- Use `plan_approval: true` for risky changes — the teammate sends a plan first, you review and approve before they write any code.
-- Let the lead coordinate. Don't micromanage teammates — they message you when done or when they're blocked.
-- After spawning, tell the user what you've set up and wait for results. Don't poll `team_status` in a loop.
+- Start with 2-3 teammates. More agents means more coordination overhead.
+- Give each teammate specific, self-contained tasks. Vague prompts produce vague results.
+- Spawn an explore agent first to understand the codebase, then spawn build agents with that context.
+- Use `worktree: false` for read-only agents (research, review, code analysis).
+- Use `plan_approval: true` for risky changes. The teammate sends a plan first, you review and approve before they write any code.
+- Don't micromanage. Teammates message you when done or when they're blocked.
+- Don't poll `team_status` in a loop. Wait for messages.
 
-### How this differs from Claude Code agent teams
+## How this differs from Claude Code agent teams
 
-Claude Code has built-in agent teams. This plugin takes the same coordination model (shared tasks, peer messaging, lead coordination) and adds:
+Same coordination model (shared tasks, peer messaging, lead coordination) with some additions:
 
-- **Git worktree isolation by default** — each teammate gets their own branch, no merge conflicts between parallel agents
-- **System prompt injection** — the lead's system prompt is updated with team state (member statuses, task progress) so it stays aware across turns
-- **Compaction safety** — team context is preserved when sessions get long and OpenCode compacts the conversation
-- **Team-aware shell environment** — teammate shells get `ENSEMBLE_TEAM`, `ENSEMBLE_MEMBER`, `ENSEMBLE_ROLE`, and `ENSEMBLE_BRANCH` variables
-- **Graceful shutdown** — teammates finish their current work before stopping, with a force flag for emergencies
-- **Plan approval mode** — review teammate plans before they write code, useful for risky or unfamiliar changes
-- **Works today as a plugin** — no waiting for upstream features, install and go
+- **Git worktree isolation by default**: each teammate gets their own branch, no merge conflicts between parallel agents
+- **System prompt injection**: the lead's system prompt is updated with team state so it stays aware across turns
+- **Compaction safety**: team context is preserved when sessions get long
+- **Team-aware shell environment**: `ENSEMBLE_TEAM`, `ENSEMBLE_MEMBER`, `ENSEMBLE_ROLE`, `ENSEMBLE_BRANCH`
+- **Graceful shutdown**: teammates finish current work before stopping, with a force flag for emergencies
+- **Plan approval mode**: review teammate plans before they write code
+- **Works today as a plugin**: install and go, no upstream changes needed
 
 ## Development
 
 ```bash
 bun install
 bun run typecheck
-bun test             # 144 tests
+bun test             # 286 tests
 bun run build
 ```
 

package/dist/index.js

@@ -13708,9 +13708,10 @@ var plugin = async (input) => {
         args: {},
         async execute(_args, ctx) {
           const result = await executeTeamStatus(deps, ctx.sessionID);
-          const members = deps.db.query("SELECT status FROM team_member WHERE team_id IN (SELECT id FROM team WHERE lead_session_id = ? OR id IN (SELECT team_id FROM team_member WHERE session_id = ?))").all(ctx.sessionID, ctx.sessionID);
-          const busy = members.filter((m) => m.status === "busy").length;
-          ctx.metadata({ title: `Team: ${members.length} members (${busy} working)` });
+          const statusMap = { busy: "working", ready: "idle", shutdown_requested: "stopping", shutdown: "done", error: "error" };
+          const members = deps.db.query("SELECT name, status FROM team_member WHERE team_id IN (SELECT id FROM team WHERE lead_session_id = ? OR id IN (SELECT team_id FROM team_member WHERE session_id = ?))").all(ctx.sessionID, ctx.sessionID);
+          const summary = members.map((m) => `${m.name}: ${statusMap[m.status] ?? m.status}`).join(", ");
+          ctx.metadata({ title: summary || "No teammates" });
           return result;
         }
       }),

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@hueyexe/opencode-ensemble",
-  "version": "0.5.0",
+  "version": "0.5.1",
   "description": "Agent teams for OpenCode — parallel agents with peer-to-peer communication, shared tasks, and coordinated execution",
   "module": "src/index.ts",
   "main": "dist/index.js",

package/src/index.ts

@@ -346,9 +346,10 @@ const plugin: Plugin = async (input) => {
         args: {},
         async execute(_args, ctx) {
           const result = await executeTeamStatus(deps, ctx.sessionID)
-          const members = deps.db.query("SELECT status FROM team_member WHERE team_id IN (SELECT id FROM team WHERE lead_session_id = ? OR id IN (SELECT team_id FROM team_member WHERE session_id = ?))").all(ctx.sessionID, ctx.sessionID) as Array<{ status: string }>
-          const busy = members.filter(m => m.status === "busy").length
-          ctx.metadata({ title: `Team: ${members.length} members (${busy} working)` })
+          const statusMap: Record<string, string> = { busy: "working", ready: "idle", shutdown_requested: "stopping", shutdown: "done", error: "error" }
+          const members = deps.db.query("SELECT name, status FROM team_member WHERE team_id IN (SELECT id FROM team WHERE lead_session_id = ? OR id IN (SELECT team_id FROM team_member WHERE session_id = ?))").all(ctx.sessionID, ctx.sessionID) as Array<{ name: string; status: string }>
+          const summary = members.map(m => `${m.name}: ${statusMap[m.status] ?? m.status}`).join(", ")
+          ctx.metadata({ title: summary || "No teammates" })
           return result
         },
       }),