opencode-swarm-plugin 0.45.6 → 0.46.0

package/examples/commands/swarm.md

@@ -57,7 +57,7 @@ $ARGUMENTS
 │     → Caught worker blocked on database schema              │
 │     → Unblocked by coordinating with upstream worker        │
 │                                                             │
-│  ✅ Delegated planning to swarm/planner subagent            │
+│  ✅ Delegated planning to swarm-planner subagent            │
 │     → Main context stayed clean (only received JSON)        │
 │     → Scaled to 7 workers without context exhaustion        │
 │                                                             │
@@ -346,7 +346,7 @@ swarm_plan_interactive(
 >
 > **DO NOT decompose inline in the coordinator thread.** This consumes massive context with file reading, CASS queries, and reasoning.
 >
-> **ALWAYS delegate to a `swarm/planner` subagent** that returns only the validated CellTree JSON.
+> **ALWAYS delegate to a `swarm-planner` subagent** that returns only the validated CellTree JSON.
 
 **❌ Don't do this (inline planning):**
 
@@ -365,9 +365,9 @@ hive_create(title="Plan: <task>", type="task", description="Decompose into subta
 # 2. Get final prompt from swarm_plan_interactive (when ready_to_decompose=true)
 # final_prompt = <from last swarm_plan_interactive call>
 
-# 3. Delegate to swarm/planner subagent
+# 3. Delegate to swarm-planner subagent
 Task(
-  subagent_type="swarm/planner",
+  subagent_type="swarm-planner",
   description="Decompose task: <task>",
   prompt="
 You are a swarm planner. Generate a CellTree for this task.
@@ -459,7 +459,7 @@ See full skill list with skills_list().
 Then spawn:
 
 ```bash
-Task(subagent_type="swarm/worker", description="<bead-title>", prompt="<from swarm_spawn_subtask>")
+Task(subagent_type="swarm-worker", description="<bead-title>", prompt="<from swarm_spawn_subtask>")
 ```
 
 **Event tracked:** `worker_spawned` (for each worker)
@@ -639,7 +639,7 @@ These events are now tracked for coordinator evaluation:
 | `session_initialized`    | swarmmail_init called                     |
 | `skill_loaded`           | skills_use called                         |
 | `researcher_spawned`     | Task(subagent_type="swarm-researcher")    |
-| `worker_spawned`         | Task(subagent_type="swarm/worker")        |
+| `worker_spawned`         | Task(subagent_type="swarm-worker")        |
 | `decomposition_complete` | hive_create_epic called                   |
 | `inbox_checked`          | swarmmail_inbox called                    |
 | `blocker_resolved`       | Coordinator unblocked stuck worker        |
@@ -672,7 +672,7 @@ Not: Do Everything Inline → Run Out of Context → Fail
 - [ ] Knowledge gathered (semantic-memory, CASS, pdf-brain, skills)
 - [ ] **Skills loaded** → Event: `skill_loaded` (per skill)
 - [ ] **Researcher spawned if needed** → Event: `researcher_spawned`
-- [ ] **Planning delegated to swarm/planner subagent** (NOT inline)
+- [ ] **Planning delegated to swarm-planner subagent** (NOT inline)
 - [ ] CellTree validated (no file conflicts)
 - [ ] Epic + subtasks created → Event: `decomposition_complete`
 - [ ] **Coordinator did NOT reserve files** (workers do this themselves)

package/global-skills/swarm-coordination/SKILL.md

@@ -310,7 +310,7 @@ Synthesize findings into `shared_context` for workers.
 >
 > **NEVER do planning inline in the coordinator thread.** Decomposition work (file reading, CASS searching, reasoning about task breakdown) consumes massive amounts of context and will exhaust your token budget on long swarms.
 >
-> **ALWAYS delegate planning to a `swarm/planner` subagent** and receive only the structured CellTree JSON result back.
+> **ALWAYS delegate planning to a `swarm-planner` subagent** and receive only the structured CellTree JSON result back.
 
 **❌ Anti-Pattern (Context-Heavy):**
 
@@ -332,9 +332,9 @@ await hive_create({
   description: `Decompose into subtasks. Context: ${synthesizedContext}`,
 });
 
-// 2. Delegate to swarm/planner subagent
+// 2. Delegate to swarm-planner subagent
 const planningResult = await Task({
-  subagent_type: "swarm/planner",
+  subagent_type: "swarm-planner",
   description: `Decompose task: ${taskTitle}`,
   prompt: `
 You are a swarm planner. Generate a CellTree for this task.
@@ -438,7 +438,7 @@ for (const subtask of subtasks) {
 
   // Spawn via Task tool
   Task({
-    subagent_type: "swarm/worker",
+    subagent_type: "swarm-worker",
     prompt: prompt.worker_prompt,
   });
 }
@@ -520,7 +520,7 @@ swarm_complete({ ... });
 
 // Coordinator immediately spawns next worker
 // ⚠️ WRONG - No quality gate!
-Task({ subagent_type: "swarm/worker", prompt: nextWorkerPrompt });
+Task({ subagent_type: "swarm-worker", prompt: nextWorkerPrompt });
 ```
 
 **✅ Correct Pattern (Review Before Proceeding):**
@@ -535,7 +535,7 @@ swarm_review({ ... });
 swarm_review_feedback({ status: "approved" });
 
 // ONLY THEN spawn next worker
-Task({ subagent_type: "swarm/worker", prompt: nextWorkerPrompt });
+Task({ subagent_type: "swarm-worker", prompt: nextWorkerPrompt });
 ```
 
 **Review Workflow (3-Strike Rule):**

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-swarm-plugin",
-  "version": "0.45.6",
+  "version": "0.46.0",
   "description": "Multi-agent swarm coordination for OpenCode with learning capabilities, beads integration, and Agent Mail",
   "type": "module",
   "main": "./dist/index.js",
@@ -43,6 +43,7 @@
     "test": "bun test --timeout 10000 src/anti-patterns.test.ts src/mandate-promotion.test.ts src/mandate-storage.test.ts src/output-guardrails.test.ts src/pattern-maturity.test.ts src/skills.test.ts src/structured.test.ts src/schemas/",
     "test:integration": "bun test --timeout 60000 src/*.integration.test.ts",
     "test:all": "bun test --timeout 60000 src/",
+    "test:isolated": "bun test ./src/zz-compaction-hook.test.ts && bun test ./src/zz-compaction-observability.integration.test.ts && bun test ./src/zz-eval-runner.test.ts",
     "test:watch": "bun test --watch src/",
     "typecheck": "tsc --noEmit",
     "eval:run": "bun --env-file=.env run bunx evalite run evals/",
@@ -63,11 +64,12 @@
     "minimatch": "^10.1.1",
     "pino": "^9.6.0",
     "pino-roll": "^1.3.0",
-    "swarm-mail": "1.6.2",
+    "swarm-mail": "1.7.0",
     "yaml": "^2.8.2",
     "zod": "4.1.8"
   },
   "devDependencies": {
+    "@libsql/client": "^0.15.15",
     "@types/bun": "latest",
     "@types/minimatch": "^6.0.0",
     "ai": "6.0.0-beta.150",
@@ -75,7 +77,8 @@
     "evalite": "^0.19.0",
     "pino-pretty": "^13.1.3",
     "turbo": "^2.6.3",
-    "typescript": "^5.7.0"
+    "typescript": "^5.7.0",
+    "vitest": "^3.0.0"
   },
   "peerDependencies": {
     "@opencode-ai/plugin": "^1.0.0"