oh-my-opencode 4.12.0 → 4.12.1

package/dist/skills/ultraresearch/SKILL.md

@@ -51,6 +51,14 @@ The research is done when all of these hold:
 - Every claim in the deliverable cites a source or a verification artifact.
 - The session journal reconstructs what was searched, found, and expanded, wave by wave.
 
+## Run the swarm as a cooperating team
+
+Saturation research is the textbook case for a cooperating team, not isolated fire-and-forget workers: a lead one worker surfaces almost always reshapes what another should search next. So when your harness gives you real cooperating members — Codex: the `teammode` skill (`codex_app` threads); OpenCode: `team_mode` — run this swarm as a team. Fall back to the background-worker swarm below only when team mode is unavailable, or the axes are genuinely independent with no cross-pollination expected.
+
+- **One member per axis — by part, ownership, or perspective, never a job title.** Each Phase 0 axis is one member owning one concrete slice: a codebase part, a source territory, or a question lens. No two members share an angle. "Backend researcher" or "the web person" gives no real boundary and invites overlap — name what the member owns.
+- **The raise law — broadcast every lead the instant it surfaces.** Members over-communicate relentlessly: every new lead, finding, contradiction, and dead end is raised to you the moment it surfaces, never hoarded for a final dump. Through long passes they send `WORKING: <axis> - <phase>`, and `BLOCKED: <reason>` the moment progress stops, so you always know a member is alive. Too many small updates is correct here; going quiet is the only failure.
+- **You lead; expand on each raised lead.** Members raise via message text, never write session files. Journal each lead and spawn its expansion the instant it lands (Phase 2), not only when a member's final reply arrives.
+
 ## Worker ground rules
 
 Research workers (explore, librarian, browsing) differ by harness, but assume:
@@ -106,7 +114,7 @@ Append each digest the moment its worker returns, not in a batch at the end —
 
 ## Phase 1 — Saturation wave
 
-Launch every first-wave worker in a single turn, all in background. Sequential launches and "start with one and see" defeat the mode.
+Launch the entire first wave in one turn — every axis at once, as team members if you formed a team, else as background workers. Sequential launches and "start with one and see" defeat the mode.
 
 Scaling floor — more angles always justify more workers:
 
@@ -137,7 +145,7 @@ End your reply with the ## EXPAND tail: '- LEAD: <discovery> — WHY: <why> —
 
 ## Phase 2 — Expand until convergence
 
-This loop is what makes the mode research rather than search. Collect workers as they finish — never wait for the full wave:
+This loop is what makes the mode research rather than search. Collect returns as they land — and in team mode, act on each lead the moment a member raises it, never waiting for the full wave or a member's final reply:
 
 1. Journal the return: digest plus verbatim EXPAND markers into `wave-<N>-<kind>-<axis>.md`.
 2. Deduplicate new markers against `expansion-log.md` — every lead ever seen, not just confirmed ones, or rejected leads resurface each wave.
@@ -221,6 +229,7 @@ High-yield combinations: official docs (`site:<docs domain>`), GitHub implementa
 | Failure | Correction |
 |---|---|
 | Sequential spawning, or trimming the first wave | All first-wave workers in one turn, background, scaling floor respected |
+| A team member hoards leads for one final dump | Raise law — every lead, finding, and dead end broadcast the moment it surfaces |
 | Worker reply without the EXPAND tail | One follow-up demanding it; the lane stays open until it lands |
 | Stopping after wave 1 because "enough was found" | Convergence rules only: 2+ expansion waves, leads run dry |
 | Obeying a surrounding "stop exploring" rule mid-research | Authority section — those rules do not bind this mode |

package/dist/tools/background-task/constants.d.ts

@@ -1,3 +1,3 @@
-export declare const BACKGROUND_TASK_DESCRIPTION = "Run agent task in background. Returns a background task ID (`bg_...`) immediately; notifies on completion.\n\nUse `background_output` to get results. Prompts MUST be in English.";
+export declare const BACKGROUND_TASK_DESCRIPTION = "Run agent task in background. Returns a background task ID (`bg_...`) immediately and notifies on completion.\n\nDo NOT poll for results. The system delivers a <system-reminder> when the task finishes.\n\nPrompts MUST be in English.";
 export declare const BACKGROUND_OUTPUT_DESCRIPTION = "Get output from background task. Use full_session=true to fetch session messages with filters. System notifies on completion, so block=true rarely needed. - Timeout values are in milliseconds (ms), NOT seconds.\n\nIMPORTANT: ONLY call this tool AFTER receiving a <system-reminder> notification for the task. Do NOT call immediately after launching a background task - wait for the notification first.";
 export declare const BACKGROUND_CANCEL_DESCRIPTION = "Cancel running background task(s). Use all=true to cancel ALL before final answer.";

package/dist/tui.js

@@ -4163,6 +4163,14 @@ var init_ast_grep = __esm(() => {
 
 // packages/utils/src/codegraph/env.ts
 var init_env = () => {};
+
+// packages/utils/src/codegraph/workspace.ts
+var init_workspace = () => {};
+
+// packages/utils/src/codegraph/guidance.ts
+var init_guidance = __esm(() => {
+  init_workspace();
+});
 // packages/utils/src/codegraph/manifest.ts
 var init_manifest = () => {};
 
@@ -4183,12 +4191,10 @@ var init_resolve2 = __esm(() => {
   requireFromHere = createRequire(import.meta.url);
 });
 
-// packages/utils/src/codegraph/workspace.ts
-var init_workspace = () => {};
-
 // packages/utils/src/codegraph/index.ts
 var init_codegraph = __esm(() => {
   init_env();
+  init_guidance();
   init_provision();
   init_resolve2();
   init_workspace();

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-opencode",
-  "version": "4.12.0",
+  "version": "4.12.1",
   "description": "The Best AI Agent Harness - Batteries-Included OpenCode Plugin with Multi-Model Orchestration, Parallel Background Agents, and Crafted LSP/AST Tools",
   "main": "./dist/index.js",
   "types": "dist/index.d.ts",
@@ -170,18 +170,18 @@
     "typescript": "^6.0.3"
   },
   "optionalDependencies": {
-    "oh-my-opencode-darwin-arm64": "4.12.0",
-    "oh-my-opencode-darwin-x64": "4.12.0",
-    "oh-my-opencode-darwin-x64-baseline": "4.12.0",
-    "oh-my-opencode-linux-arm64": "4.12.0",
-    "oh-my-opencode-linux-arm64-musl": "4.12.0",
-    "oh-my-opencode-linux-x64": "4.12.0",
-    "oh-my-opencode-linux-x64-baseline": "4.12.0",
-    "oh-my-opencode-linux-x64-musl": "4.12.0",
-    "oh-my-opencode-linux-x64-musl-baseline": "4.12.0",
-    "oh-my-opencode-windows-arm64": "4.12.0",
-    "oh-my-opencode-windows-x64": "4.12.0",
-    "oh-my-opencode-windows-x64-baseline": "4.12.0"
+    "oh-my-opencode-darwin-arm64": "4.12.1",
+    "oh-my-opencode-darwin-x64": "4.12.1",
+    "oh-my-opencode-darwin-x64-baseline": "4.12.1",
+    "oh-my-opencode-linux-arm64": "4.12.1",
+    "oh-my-opencode-linux-arm64-musl": "4.12.1",
+    "oh-my-opencode-linux-x64": "4.12.1",
+    "oh-my-opencode-linux-x64-baseline": "4.12.1",
+    "oh-my-opencode-linux-x64-musl": "4.12.1",
+    "oh-my-opencode-linux-x64-musl-baseline": "4.12.1",
+    "oh-my-opencode-windows-arm64": "4.12.1",
+    "oh-my-opencode-windows-x64": "4.12.1",
+    "oh-my-opencode-windows-x64-baseline": "4.12.1"
   },
   "overrides": {
     "hono": "^4.12.18",

package/packages/omo-codex/plugin/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "omo",
-  "version": "4.12.0",
+  "version": "4.12.1",
   "description": "One Codex plugin namespace for Yeongyu's local Codex components.",
   "author": {
     "name": "Yeongyu Kim",
@@ -32,7 +32,9 @@
     "./hooks/pre-tool-use-enforcing-unlimited-goal-budget.json",
     "./hooks/post-tool-use-checking-comments.json",
     "./hooks/post-tool-use-checking-lsp-diagnostics.json",
+    "./hooks/post-tool-use-checking-codegraph-init-guidance.json",
     "./hooks/post-tool-use-matching-project-rules.json",
+    "./hooks/post-tool-use-checking-thread-title-hygiene.json",
     "./hooks/post-compact-resetting-git-bash-mcp-reminder.json",
     "./hooks/post-compact-resetting-project-rule-cache.json",
     "./hooks/post-compact-resetting-lsp-diagnostics-cache.json",

package/packages/omo-codex/plugin/components/bootstrap/package.json

@@ -1,6 +1,6 @@
 {
 	"name": "@sisyphuslabs/codex-bootstrap",
-	"version": "4.12.0",
+	"version": "4.12.1",
 	"description": "Codex SessionStart bootstrap component that provisions LazyCodex runtime dependencies from a detached worker.",
 	"type": "module",
 	"private": true,