oh-my-codex 0.17.0 → 0.17.2

package/plugins/oh-my-codex/.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "oh-my-codex",
-  "version": "0.17.0",
+  "version": "0.17.2",
   "description": "Multi-agent orchestration layer for OpenAI Codex CLI",
   "author": {
     "name": "Yeachan Heo",

package/plugins/oh-my-codex/skills/plan/SKILL.md

@@ -95,8 +95,8 @@ Jumping into code without understanding requirements leads to rework, scope cree
    c. Update the plan file in `.omx/plans/` with the accepted improvements (add missing details, refine steps, strengthen acceptance criteria, ADR updates, etc.)
    d. Note which improvements were applied in a brief changelog section at the end of the plan
    e. Before any execution handoff, derive an explicit **available-agent-types roster** from the known prompt catalog and add concrete **follow-up staffing guidance** for both `$ralph` and `$team` (recommended roles, counts, suggested reasoning levels by lane, and why each lane exists)
-   f. Add a product-facing **Goal-Mode Follow-up Suggestions** section: recommend `$ultragoal` by default for general goal-oriented follow-up, `$autoresearch-goal` when the context is a research project, and `$performance-goal` when the context is an optimization or performance project. Keep these suggestions alongside the Ralph/team paths rather than replacing them when implementation delivery is still the main need.
-   g. For the `$team` path, add an explicit launch-hint block with concrete `omx team` / `$team` commands and a **team verification path** (what team proves before shutdown, what Ralph verifies after handoff)
+   f. Add a product-facing **Goal-Mode Follow-up Suggestions** section: recommend `$ultragoal` by default for general goal-oriented follow-up, `$autoresearch-goal` when the context is a research project, and `$performance-goal` when the context is an optimization or performance project. Keep these suggestions alongside the Ralph/team paths rather than replacing them when implementation delivery is still the main need. For durable-goal work that is also parallelizable, explicitly recommend **Team + Ultragoal**: Ultragoal remains leader-owned goal/ledger state and Team returns checkpoint-ready execution evidence.
+   g. For the `$team` path, add an explicit launch-hint block with concrete `omx team` / `$team` commands and a **team verification path** (what team proves before shutdown, what Ralph verifies after handoff). Distinguish Team + Ultragoal from a later Ralph follow-up: Team handles coordinated parallel lanes; Ralph is only for persistent sequential single-owner verification/fix pressure when needed.
 7. On Critic approval (with improvements applied): *(--interactive only)* If running with `--interactive`, use `AskUserQuestion` / the structured question UI to present the plan with these options:
    - **Approve and execute** — proceed to implementation via ralph+ultrawork
    - **Approve and implement via team** — proceed to implementation via coordinated parallel team agents
@@ -148,7 +148,7 @@ Plans are saved to `.omx/plans/`. Drafts go to `.omx/drafts/`.
 - In consensus mode, default to RALPLAN-DR short mode; enable deliberate mode on `--deliberate` or explicit high-risk signals (auth/security, migrations, destructive changes, production incidents, compliance/PII, public API breakage)
 - In consensus mode with `--interactive`: use `AskUserQuestion` / the structured question UI for the user feedback step (step 2) and the final approval step (step 7) -- never ask for approval in plain text when a structured surface is available. Without `--interactive`, auto-proceed through planning steps without pausing. Output the final plan without execution.
 - In consensus mode with `--interactive`, on user approval **MUST** invoke the selected follow-up lane from step 9 (`$ralph`, `$team`, `$ultragoal`, `$autoresearch-goal`, or `$performance-goal`) -- never implement directly in the planning agent
-- In consensus mode, execution follow-up handoff **MUST** include an explicit available-agent-types roster plus concrete staffing / role-allocation guidance grounded in that roster, suggested reasoning levels by lane, product-facing goal-mode follow-up suggestions (`$ultragoal` by default, `$autoresearch-goal` for research projects, `$performance-goal` for optimization/performance projects), explicit `omx team` / `$team` launch hints, and a team verification path
+- In consensus mode, execution follow-up handoff **MUST** include an explicit available-agent-types roster plus concrete staffing / role-allocation guidance grounded in that roster, suggested reasoning levels by lane, product-facing goal-mode follow-up suggestions (`$ultragoal` by default, `$autoresearch-goal` for research projects, `$performance-goal` for optimization/performance projects), explicit `omx team` / `$team` launch hints, and a team verification path. For parallelizable durable-goal plans, recommend Team + Ultragoal with leader-owned checkpointing from Team evidence; reserve Ralph for persistent sequential single-owner verification/fix follow-up.
 </Tool_Usage>
 
 ## Scenario Examples

package/plugins/oh-my-codex/skills/ralph/SKILL.md

@@ -125,8 +125,8 @@ Use the CLI-first state surface for Ralph lifecycle state (`omx state write/read
   `omx state write --input '{"mode":"ralph","iteration":<current>,"current_phase":"executing"}' --json`
 - **On verification/fix transition**:
   `omx state write --input '{"mode":"ralph","current_phase":"verifying"}' --json` or `omx state write --input '{"mode":"ralph","current_phase":"fixing"}' --json`
-- **On completion**:
-  `omx state write --input '{"mode":"ralph","active":false,"current_phase":"complete","completed_at":"<now>"}' --json`
+- **On completion** (only after the completion audit passes with real evidence):
+  `omx state write --input '{"mode":"ralph","active":false,"current_phase":"complete","completed_at":"<now>","completion_audit":{"passed":true,"prompt_to_artifact_checklist":["<requirement mapped to artifact/evidence>"],"verification_evidence":["<fresh test/build/lint command and result>"]}}' --json`
 - **On cancellation/cleanup**:
   run `$cancel` (which should call `omx state clear --input '{"mode":"ralph"}' --json`)
 

package/plugins/oh-my-codex/skills/ralplan/SKILL.md

@@ -70,7 +70,7 @@ When ralplan outputs a final handoff or asks the user to choose a next lane, inc
 - `$autoresearch-goal` — research-project follow-up when the plan centers on a question, literature/reference gathering, evaluator-backed research, or a professor/critic-style research deliverable.
 - `$performance-goal` — optimization/performance follow-up when the plan centers on speed, latency, throughput, memory, benchmark, or other measurable performance work.
 
-Keep `$ralph` and `$team` as first-class execution options where appropriate: use Ralph for persistent single-owner completion/verification pressure and team for coordinated parallel implementation. Do not present the goal-mode options as replacements for Ralph/team when the task is mainly implementation delivery; present them as better fits when durable goal tracking, research validation, or performance evaluators are the primary need.
+Keep `$ralph` and `$team` as first-class execution options where appropriate: use Ralph for persistent single-owner completion/verification pressure and team for coordinated parallel implementation. For parallelizable durable-goal delivery, recommend `$ultragoal` + `$team` together: Ultragoal remains the leader-owned `.omx/ultragoal` ledger/Codex-goal wrapper while Team runs parallel lanes and returns checkpoint-ready evidence. Do not present the goal-mode options as replacements for Ralph/team when the task is mainly implementation delivery; present them as better fits when durable goal tracking, research validation, or performance evaluators are the primary need.
 
 ## Pre-context Intake
 

package/plugins/oh-my-codex/skills/team/SKILL.md

@@ -57,6 +57,12 @@ requiring a separate linked Ralph launch up front.
 - **Escalation:** start a separate `omx ralph ...` / `$ralph ...` only when a later manual follow-up still needs a persistent single-owner fix/verification loop.
 - **Deprecation:** `omx team ralph ...` has been removed. Use plain `omx team ...` for team execution or run `omx ralph ...` separately when you explicitly want a later Ralph loop.
 
+### Team + Ultragoal bridge
+
+Use `$ultragoal` for durable leader-owned goal/ledger tracking and `$team` for parallel execution lanes. When Team is launched with an active `.omx/ultragoal/goals.json`, worker inboxes/status may include leader-owned Ultragoal context: `.omx/ultragoal/goals.json`, `.omx/ultragoal/ledger.jsonl`, the active goal id, Codex goal mode, and the `fresh_leader_get_goal_required` checkpoint policy.
+
+Workers provide task status and verification evidence only. They do not own Ultragoal goal state, create worker ledgers, mutate `.omx/ultragoal`, auto-launch Team from Ultragoal, or perform hidden Codex goal mutation. The leader uses terminal Team evidence plus a fresh `get_goal` snapshot to run `omx ultragoal checkpoint --goal-id <id> --status complete --evidence "<team evidence mentioning .omx/ultragoal and <id>>" --codex-goal-json <fresh-get_goal-json-or-path>`.
+
 ### Claude teammates (v0.6.0+)
 
 Important: `N:agent-type` (for example `2:executor`) selects the **worker role prompt**, not the worker CLI (`codex` vs `claude`).

package/plugins/oh-my-codex/skills/ultragoal/SKILL.md

@@ -43,6 +43,17 @@ Loop until `omx ultragoal status` reports all goals complete:
    `omx ultragoal checkpoint --goal-id <id> --status blocked --evidence "<completed legacy Codex goal blocks create_goal in this thread>" --codex-goal-json <get_goal-json-or-path>`
 11. Resume failed goals with `omx ultragoal complete-goals --retry-failed`.
 
+## Use Ultragoal and Team together
+
+Use ultragoal and team together for a durable Ultragoal story that benefits from parallel execution. Ultragoal remains leader-owned: `.omx/ultragoal/goals.json` stores the story plan and `.omx/ultragoal/ledger.jsonl` stores checkpoints. Team is the parallel execution engine and returns task/evidence status to the leader.
+
+The leader checkpoints Ultragoal from Team evidence with a fresh `get_goal` snapshot:
+
+```sh
+omx ultragoal checkpoint --goal-id <id> --status complete --evidence "<team evidence mentioning .omx/ultragoal and <id>>" --codex-goal-json <fresh-get_goal-json-or-path>
+```
+
+Workers do not own ultragoal goal state, do not create worker ultragoal ledgers, and do not checkpoint Ultragoal. Team launch remains explicit; Ultragoal does not auto-launch Team and performs no hidden Codex goal mutation.
 
 ## Mandatory final cleanup and review gate
 

package/skills/plan/SKILL.md

@@ -95,8 +95,8 @@ Jumping into code without understanding requirements leads to rework, scope cree
    c. Update the plan file in `.omx/plans/` with the accepted improvements (add missing details, refine steps, strengthen acceptance criteria, ADR updates, etc.)
    d. Note which improvements were applied in a brief changelog section at the end of the plan
    e. Before any execution handoff, derive an explicit **available-agent-types roster** from the known prompt catalog and add concrete **follow-up staffing guidance** for both `$ralph` and `$team` (recommended roles, counts, suggested reasoning levels by lane, and why each lane exists)
-   f. Add a product-facing **Goal-Mode Follow-up Suggestions** section: recommend `$ultragoal` by default for general goal-oriented follow-up, `$autoresearch-goal` when the context is a research project, and `$performance-goal` when the context is an optimization or performance project. Keep these suggestions alongside the Ralph/team paths rather than replacing them when implementation delivery is still the main need.
-   g. For the `$team` path, add an explicit launch-hint block with concrete `omx team` / `$team` commands and a **team verification path** (what team proves before shutdown, what Ralph verifies after handoff)
+   f. Add a product-facing **Goal-Mode Follow-up Suggestions** section: recommend `$ultragoal` by default for general goal-oriented follow-up, `$autoresearch-goal` when the context is a research project, and `$performance-goal` when the context is an optimization or performance project. Keep these suggestions alongside the Ralph/team paths rather than replacing them when implementation delivery is still the main need. For durable-goal work that is also parallelizable, explicitly recommend **Team + Ultragoal**: Ultragoal remains leader-owned goal/ledger state and Team returns checkpoint-ready execution evidence.
+   g. For the `$team` path, add an explicit launch-hint block with concrete `omx team` / `$team` commands and a **team verification path** (what team proves before shutdown, what Ralph verifies after handoff). Distinguish Team + Ultragoal from a later Ralph follow-up: Team handles coordinated parallel lanes; Ralph is only for persistent sequential single-owner verification/fix pressure when needed.
 7. On Critic approval (with improvements applied): *(--interactive only)* If running with `--interactive`, use `AskUserQuestion` / the structured question UI to present the plan with these options:
    - **Approve and execute** — proceed to implementation via ralph+ultrawork
    - **Approve and implement via team** — proceed to implementation via coordinated parallel team agents
@@ -148,7 +148,7 @@ Plans are saved to `.omx/plans/`. Drafts go to `.omx/drafts/`.
 - In consensus mode, default to RALPLAN-DR short mode; enable deliberate mode on `--deliberate` or explicit high-risk signals (auth/security, migrations, destructive changes, production incidents, compliance/PII, public API breakage)
 - In consensus mode with `--interactive`: use `AskUserQuestion` / the structured question UI for the user feedback step (step 2) and the final approval step (step 7) -- never ask for approval in plain text when a structured surface is available. Without `--interactive`, auto-proceed through planning steps without pausing. Output the final plan without execution.
 - In consensus mode with `--interactive`, on user approval **MUST** invoke the selected follow-up lane from step 9 (`$ralph`, `$team`, `$ultragoal`, `$autoresearch-goal`, or `$performance-goal`) -- never implement directly in the planning agent
-- In consensus mode, execution follow-up handoff **MUST** include an explicit available-agent-types roster plus concrete staffing / role-allocation guidance grounded in that roster, suggested reasoning levels by lane, product-facing goal-mode follow-up suggestions (`$ultragoal` by default, `$autoresearch-goal` for research projects, `$performance-goal` for optimization/performance projects), explicit `omx team` / `$team` launch hints, and a team verification path
+- In consensus mode, execution follow-up handoff **MUST** include an explicit available-agent-types roster plus concrete staffing / role-allocation guidance grounded in that roster, suggested reasoning levels by lane, product-facing goal-mode follow-up suggestions (`$ultragoal` by default, `$autoresearch-goal` for research projects, `$performance-goal` for optimization/performance projects), explicit `omx team` / `$team` launch hints, and a team verification path. For parallelizable durable-goal plans, recommend Team + Ultragoal with leader-owned checkpointing from Team evidence; reserve Ralph for persistent sequential single-owner verification/fix follow-up.
 </Tool_Usage>
 
 ## Scenario Examples

package/skills/ralph/SKILL.md

@@ -125,8 +125,8 @@ Use the CLI-first state surface for Ralph lifecycle state (`omx state write/read
   `omx state write --input '{"mode":"ralph","iteration":<current>,"current_phase":"executing"}' --json`
 - **On verification/fix transition**:
   `omx state write --input '{"mode":"ralph","current_phase":"verifying"}' --json` or `omx state write --input '{"mode":"ralph","current_phase":"fixing"}' --json`
-- **On completion**:
-  `omx state write --input '{"mode":"ralph","active":false,"current_phase":"complete","completed_at":"<now>"}' --json`
+- **On completion** (only after the completion audit passes with real evidence):
+  `omx state write --input '{"mode":"ralph","active":false,"current_phase":"complete","completed_at":"<now>","completion_audit":{"passed":true,"prompt_to_artifact_checklist":["<requirement mapped to artifact/evidence>"],"verification_evidence":["<fresh test/build/lint command and result>"]}}' --json`
 - **On cancellation/cleanup**:
   run `$cancel` (which should call `omx state clear --input '{"mode":"ralph"}' --json`)
 

package/skills/ralplan/SKILL.md

@@ -70,7 +70,7 @@ When ralplan outputs a final handoff or asks the user to choose a next lane, inc
 - `$autoresearch-goal` — research-project follow-up when the plan centers on a question, literature/reference gathering, evaluator-backed research, or a professor/critic-style research deliverable.
 - `$performance-goal` — optimization/performance follow-up when the plan centers on speed, latency, throughput, memory, benchmark, or other measurable performance work.
 
-Keep `$ralph` and `$team` as first-class execution options where appropriate: use Ralph for persistent single-owner completion/verification pressure and team for coordinated parallel implementation. Do not present the goal-mode options as replacements for Ralph/team when the task is mainly implementation delivery; present them as better fits when durable goal tracking, research validation, or performance evaluators are the primary need.
+Keep `$ralph` and `$team` as first-class execution options where appropriate: use Ralph for persistent single-owner completion/verification pressure and team for coordinated parallel implementation. For parallelizable durable-goal delivery, recommend `$ultragoal` + `$team` together: Ultragoal remains the leader-owned `.omx/ultragoal` ledger/Codex-goal wrapper while Team runs parallel lanes and returns checkpoint-ready evidence. Do not present the goal-mode options as replacements for Ralph/team when the task is mainly implementation delivery; present them as better fits when durable goal tracking, research validation, or performance evaluators are the primary need.
 
 ## Pre-context Intake
 

package/skills/team/SKILL.md

@@ -57,6 +57,12 @@ requiring a separate linked Ralph launch up front.
 - **Escalation:** start a separate `omx ralph ...` / `$ralph ...` only when a later manual follow-up still needs a persistent single-owner fix/verification loop.
 - **Deprecation:** `omx team ralph ...` has been removed. Use plain `omx team ...` for team execution or run `omx ralph ...` separately when you explicitly want a later Ralph loop.
 
+### Team + Ultragoal bridge
+
+Use `$ultragoal` for durable leader-owned goal/ledger tracking and `$team` for parallel execution lanes. When Team is launched with an active `.omx/ultragoal/goals.json`, worker inboxes/status may include leader-owned Ultragoal context: `.omx/ultragoal/goals.json`, `.omx/ultragoal/ledger.jsonl`, the active goal id, Codex goal mode, and the `fresh_leader_get_goal_required` checkpoint policy.
+
+Workers provide task status and verification evidence only. They do not own Ultragoal goal state, create worker ledgers, mutate `.omx/ultragoal`, auto-launch Team from Ultragoal, or perform hidden Codex goal mutation. The leader uses terminal Team evidence plus a fresh `get_goal` snapshot to run `omx ultragoal checkpoint --goal-id <id> --status complete --evidence "<team evidence mentioning .omx/ultragoal and <id>>" --codex-goal-json <fresh-get_goal-json-or-path>`.
+
 ### Claude teammates (v0.6.0+)
 
 Important: `N:agent-type` (for example `2:executor`) selects the **worker role prompt**, not the worker CLI (`codex` vs `claude`).

package/skills/ultragoal/SKILL.md

@@ -43,6 +43,17 @@ Loop until `omx ultragoal status` reports all goals complete:
    `omx ultragoal checkpoint --goal-id <id> --status blocked --evidence "<completed legacy Codex goal blocks create_goal in this thread>" --codex-goal-json <get_goal-json-or-path>`
 11. Resume failed goals with `omx ultragoal complete-goals --retry-failed`.
 
+## Use Ultragoal and Team together
+
+Use ultragoal and team together for a durable Ultragoal story that benefits from parallel execution. Ultragoal remains leader-owned: `.omx/ultragoal/goals.json` stores the story plan and `.omx/ultragoal/ledger.jsonl` stores checkpoints. Team is the parallel execution engine and returns task/evidence status to the leader.
+
+The leader checkpoints Ultragoal from Team evidence with a fresh `get_goal` snapshot:
+
+```sh
+omx ultragoal checkpoint --goal-id <id> --status complete --evidence "<team evidence mentioning .omx/ultragoal and <id>>" --codex-goal-json <fresh-get_goal-json-or-path>
+```
+
+Workers do not own ultragoal goal state, do not create worker ultragoal ledgers, and do not checkpoint Ultragoal. Team launch remains explicit; Ultragoal does not auto-launch Team and performs no hidden Codex goal mutation.
 
 ## Mandatory final cleanup and review gate
 

package/src/scripts/__tests__/codex-native-hook.test.ts

@@ -200,6 +200,8 @@ const TEAM_ENV_KEYS = [
   "TMUX",
   "TMUX_PANE",
   "OMX_TMUX_HUD_OWNER",
+  "OMX_NATIVE_STOP_NO_PROGRESS_MAX_REPEATS",
+  "OMX_NATIVE_STOP_NO_PROGRESS_IDLE_MS",
 ] as const;
 
 const priorTeamEnv = new Map<(typeof TEAM_ENV_KEYS)[number], string | undefined>();
@@ -4403,6 +4405,87 @@ exit 0
     }
   });
 
+  it("stays silent on PreToolUse for compact inline Lore commit with only OmX co-author trailer", async () => {
+    const cwd = await mkdtemp(join(tmpdir(), "omx-native-hook-pretool-git-commit-compact-coauthor-"));
+    try {
+      const result = await dispatchCodexNativeHook(
+        {
+          hook_event_name: "PreToolUse",
+          cwd,
+          tool_name: "Bash",
+          tool_use_id: "tool-git-commit-compact-coauthor",
+          tool_input: {
+            command: [
+              'git commit',
+              '-m "Launch lvisai.xyz intro site"',
+              '-m "Co-authored-by: OmX <omx@oh-my-codex.dev>"',
+            ].join(" "),
+          },
+        },
+        { cwd },
+      );
+
+      assert.equal(result.omxEventName, "pre-tool-use");
+      assert.equal(result.outputJson, null);
+    } finally {
+      await rm(cwd, { recursive: true, force: true });
+    }
+  });
+
+  it("stays silent on PreToolUse for body-omitted inline Lore commit with decision trailers", async () => {
+    const cwd = await mkdtemp(join(tmpdir(), "omx-native-hook-pretool-git-commit-compact-trailers-"));
+    try {
+      const result = await dispatchCodexNativeHook(
+        {
+          hook_event_name: "PreToolUse",
+          cwd,
+          tool_name: "Bash",
+          tool_use_id: "tool-git-commit-compact-trailers",
+          tool_input: {
+            command: [
+              'git commit',
+              '-m "Launch lvisai.xyz intro site"',
+              '-m "Constraint: Native PreToolUse can only inspect inline Bash command text\nTested: node --test dist/scripts/__tests__/codex-native-hook.test.js\n\nCo-authored-by: OmX <omx@oh-my-codex.dev>"',
+            ].join(" "),
+          },
+        },
+        { cwd },
+      );
+
+      assert.equal(result.omxEventName, "pre-tool-use");
+      assert.equal(result.outputJson, null);
+    } finally {
+      await rm(cwd, { recursive: true, force: true });
+    }
+  });
+
+  it("blocks PreToolUse compact inline Lore commit when the blank separator is missing", async () => {
+    const cwd = await mkdtemp(join(tmpdir(), "omx-native-hook-pretool-git-commit-compact-no-separator-"));
+    try {
+      const result = await dispatchCodexNativeHook(
+        {
+          hook_event_name: "PreToolUse",
+          cwd,
+          tool_name: "Bash",
+          tool_use_id: "tool-git-commit-compact-no-separator",
+          tool_input: {
+            command: [
+              'git commit',
+              '--message="Launch lvisai.xyz intro site\nCo-authored-by: OmX <omx@oh-my-codex.dev>"',
+            ].join(" "),
+          },
+        },
+        { cwd },
+      );
+
+      assert.equal(result.omxEventName, "pre-tool-use");
+      assert.equal((result.outputJson as { decision?: string } | null)?.decision, "block");
+      assert.match(JSON.stringify(result.outputJson), /Add a blank line after the subject/);
+    } finally {
+      await rm(cwd, { recursive: true, force: true });
+    }
+  });
+
   it("warns on PreToolUse git commit when mapped source changes lack staged docs refresh", async () => {
     const cwd = await mkdtemp(join(tmpdir(), "omx-native-hook-pretool-document-refresh-warn-"));
     try {
@@ -7663,7 +7746,11 @@ exit 0
       );
 
       assert.equal(result.omxEventName, "stop");
-      assert.match(String(result.outputJson?.reason), /Ralph completion audit is missing required evidence/);
+      const reason = String(result.outputJson?.reason);
+      assert.match(reason, /Ralph completion audit is missing required evidence/);
+      assert.match(reason, /state\.completion_audit = \{ passed: true, prompt_to_artifact_checklist: \[\.\.\.\], verification_evidence: \[\.\.\.\] \}/);
+      assert.match(reason, /repo-relative JSON file/);
+      assert.match(reason, /Markdown artifacts and flat top-level checklist\/evidence fields are not accepted/);
       assert.equal(result.outputJson?.stopReason, "ralph_completion_audit_missing_completion_audit");
       const reopened = JSON.parse(await readFile(statePath, "utf-8")) as Record<string, unknown>;
       assert.equal(reopened.active, true);
@@ -8519,6 +8606,51 @@ exit 0
     }
   });
 
+  it("bounds repeated ordinary working Stop loops with a diagnostic summary", async () => {
+    const cwd = await mkdtemp(join(tmpdir(), "omx-native-hook-working-loop-"));
+    try {
+      await mkdir(join(cwd, ".omx", "state"), { recursive: true });
+      process.env.OMX_SESSION_ID = "sess-working-loop";
+      process.env.OMX_NATIVE_STOP_NO_PROGRESS_MAX_REPEATS = "2";
+      process.env.OMX_NATIVE_STOP_NO_PROGRESS_IDLE_MS = "0";
+
+      const payload = {
+        hook_event_name: "Stop",
+        cwd,
+        session_id: "sess-working-loop",
+        thread_id: "thread-working-loop",
+        turn_id: "turn-working-loop-1",
+        last_assistant_message: "Keep going and finish the cleanup.",
+      };
+
+      const first = await dispatchCodexNativeHook(payload, { cwd });
+      assert.equal(first.outputJson?.stopReason, "auto_nudge");
+
+      const repeated = await dispatchCodexNativeHook(
+        {
+          ...payload,
+          turn_id: "turn-working-loop-2",
+          stop_hook_active: true,
+        },
+        { cwd },
+      );
+
+      assert.equal(repeated.omxEventName, "stop");
+      assert.equal(repeated.outputJson?.decision, "block");
+      assert.equal(repeated.outputJson?.stopReason, "ordinary_task_no_progress_guard");
+      assert.match(String(repeated.outputJson?.systemMessage), /no-progress guard triggered/);
+      assert.match(String(repeated.outputJson?.systemMessage), /diagnostic summary/);
+      assert.match(String(repeated.outputJson?.systemMessage), /complete, blocked, failed, or needs missing information/);
+
+      const persisted = JSON.parse(
+        await readFile(join(cwd, ".omx", "state", "native-stop-state.json"), "utf-8"),
+      ) as { sessions: Record<string, { ordinary_no_progress_guard?: { repeat_count?: number } }> };
+      assert.equal(persisted.sessions["sess-working-loop"]?.ordinary_no_progress_guard?.repeat_count, 2);
+    } finally {
+      await rm(cwd, { recursive: true, force: true });
+    }
+  });
+
   it("re-blocks duplicate native auto-nudge replays for the same Stop reply", async () => {
     const cwd = await mkdtemp(join(tmpdir(), "omx-native-hook-auto-nudge-once-"));
     try {

package/src/scripts/__tests__/run-test-files.test.ts

@@ -47,6 +47,38 @@ describe('run-test-files diagnostics', () => {
     }
   });
 
+
+  it('can force-exit Node test runner after successful CI tests to avoid leaked-handle hangs', () => {
+    const wd = mkdtempSync(join(tmpdir(), 'omx-run-test-files-'));
+    try {
+      const testsDir = join(wd, '__tests__');
+      mkdirSync(testsDir, { recursive: true });
+      writeFileSync(
+        join(testsDir, 'leaky-pass.test.js'),
+        [
+          "import { test } from 'node:test';",
+          "test('passes but leaves an interval', () => { setInterval(() => {}, 1_000); });",
+          '',
+        ].join('\n'),
+      );
+
+      const withoutForceExit = runCompiledRunner(wd, { OMX_NODE_TEST_RUNNER_TIMEOUT_MS: '750' }, 2_000);
+      assert.notEqual(withoutForceExit.status, 0);
+      assert.match(withoutForceExit.stderr, /force exit disabled/);
+      assert.match(withoutForceExit.stderr, /did not exit normally|runner timeout 750ms/);
+
+      const withForceExit = runCompiledRunner(
+        wd,
+        { OMX_NODE_TEST_RUNNER_TIMEOUT_MS: '750', OMX_NODE_TEST_FORCE_EXIT: '1' },
+        2_000,
+      );
+      assert.equal(withForceExit.status, 0, withForceExit.stderr || withForceExit.stdout);
+      assert.match(withForceExit.stderr, /force exit enabled/);
+    } finally {
+      rmSync(wd, { recursive: true, force: true });
+    }
+  });
+
   it('logs that per-test timeout is disabled by default', () => {
     const wd = mkdtempSync(join(tmpdir(), 'omx-run-test-files-'));
     try {

package/src/scripts/codex-native-hook.ts

@@ -127,6 +127,9 @@ const SKILL_STOP_BLOCKERS = new Set(["ralplan"]);
 const TEAM_STOP_BLOCKING_TASK_STATUSES = new Set(["pending", "in_progress", "blocked"]);
 const TEAM_WORKER_TERMINAL_RUN_STATES = new Set(["done", "complete", "completed", "failed", "stopped", "cancelled"]);
 const NATIVE_STOP_STATE_FILE = "native-stop-state.json";
+const ORDINARY_STOP_NO_PROGRESS_DEFAULT_MAX_REPEATS = 8;
+const ORDINARY_STOP_NO_PROGRESS_DEFAULT_IDLE_MS = 10 * 60_000;
+const ORDINARY_STOP_NO_PROGRESS_MAX_MESSAGE_LENGTH = 240;
 const STABLE_FINAL_RECOMMENDATION_PATTERNS = [
   /^\s*(?:launch|release|ship)-?ready\s*:\s*(?:yes|no)\b[^\n\r]*/im,
   /^\s*ready to release\s*:\s*(?:yes|no)\b[^\n\r]*/im,
@@ -2350,6 +2353,109 @@ function readPreviousNativeStopSignature(
   return safeString(sessionState.last_signature).trim();
 }
 
+function parseBoundedPositiveInteger(value: unknown, fallback: number): number {
+  const parsed = Math.trunc(Number(value));
+  return Number.isFinite(parsed) && parsed > 0 ? parsed : fallback;
+}
+
+function parseBoundedNonNegativeInteger(value: unknown, fallback: number): number {
+  const parsed = Math.trunc(Number(value));
+  return Number.isFinite(parsed) && parsed >= 0 ? parsed : fallback;
+}
+
+function normalizeOrdinaryStopProgressText(value: unknown): string {
+  return safeString(value)
+    .replace(/\s+/g, " ")
+    .trim()
+    .toLowerCase();
+}
+
+function shortenOrdinaryStopProgressText(value: string): string {
+  const trimmed = value.replace(/\s+/g, " ").trim();
+  if (trimmed.length <= ORDINARY_STOP_NO_PROGRESS_MAX_MESSAGE_LENGTH) return trimmed;
+  return `${trimmed.slice(0, ORDINARY_STOP_NO_PROGRESS_MAX_MESSAGE_LENGTH - 1).trimEnd()}…`;
+}
+
+function ordinaryStopProgressFingerprint(payload: CodexHookPayload): string {
+  const message = normalizeOrdinaryStopProgressText(
+    payload.last_assistant_message ?? payload.lastAssistantMessage,
+  ) || "<no assistant message>";
+  const mode = normalizeOrdinaryStopProgressText(payload.mode) || "ordinary";
+  return `${mode}|${message}`;
+}
+
+function readIsoTimeMs(value: unknown): number | null {
+  const parsed = Date.parse(safeString(value));
+  return Number.isFinite(parsed) ? parsed : null;
+}
+
+async function maybeBuildOrdinaryStopNoProgressOutput(
+  payload: CodexHookPayload,
+  stateDir: string,
+  canonicalSessionId?: string,
+): Promise<Record<string, unknown> | null> {
+  const statePath = join(stateDir, NATIVE_STOP_STATE_FILE);
+  const state = await readJsonIfExists(statePath) ?? {};
+  const sessions = safeObject(state.sessions);
+  const sessionKey = readNativeStopSessionKey(payload, canonicalSessionId);
+  const sessionState = safeObject(sessions[sessionKey]);
+  const previousGuard = safeObject(sessionState.ordinary_no_progress_guard);
+  const fingerprint = ordinaryStopProgressFingerprint(payload);
+  const nowIso = new Date().toISOString();
+  const previousFingerprint = safeString(previousGuard.fingerprint).trim();
+  const sameFingerprint = previousFingerprint === fingerprint;
+  const firstSeenAt = sameFingerprint
+    ? safeString(previousGuard.first_seen_at).trim() || nowIso
+    : nowIso;
+  const repeatCount = sameFingerprint
+    ? parseBoundedPositiveInteger(previousGuard.repeat_count, 1) + 1
+    : 1;
+
+  sessions[sessionKey] = {
+    ...sessionState,
+    ordinary_no_progress_guard: {
+      fingerprint,
+      first_seen_at: firstSeenAt,
+      last_seen_at: nowIso,
+      repeat_count: repeatCount,
+      last_turn_id: readPayloadTurnId(payload) || null,
+      last_thread_id: readPayloadThreadId(payload) || null,
+    },
+  };
+  await mkdir(stateDir, { recursive: true });
+  await writeFile(statePath, JSON.stringify({ ...state, sessions }, null, 2));
+
+  const stopHookActive = payload.stop_hook_active === true || payload.stopHookActive === true;
+  if (!stopHookActive) return null;
+
+  const maxRepeats = parseBoundedPositiveInteger(
+    process.env.OMX_NATIVE_STOP_NO_PROGRESS_MAX_REPEATS,
+    ORDINARY_STOP_NO_PROGRESS_DEFAULT_MAX_REPEATS,
+  );
+  const idleMs = parseBoundedNonNegativeInteger(
+    process.env.OMX_NATIVE_STOP_NO_PROGRESS_IDLE_MS,
+    ORDINARY_STOP_NO_PROGRESS_DEFAULT_IDLE_MS,
+  );
+  const firstSeenMs = readIsoTimeMs(firstSeenAt) ?? Date.now();
+  const elapsedMs = Math.max(0, Date.now() - firstSeenMs);
+  if (repeatCount < maxRepeats || elapsedMs < idleMs) return null;
+
+  const message = shortenOrdinaryStopProgressText(
+    safeString(payload.last_assistant_message ?? payload.lastAssistantMessage) || "no assistant message recorded",
+  );
+  const elapsedSeconds = Math.round(elapsedMs / 1000);
+  const diagnostic =
+    `OMX ordinary task no-progress guard triggered after ${repeatCount} repeated Stop-hook pass(es) over ~${elapsedSeconds}s with unchanged status: "${message}". ` +
+    "Emit a concise diagnostic summary now: state the last concrete progress/evidence, whether the task is complete, blocked, failed, or needs missing information, and stop instead of continuing a vague working loop.";
+
+  return {
+    decision: "block",
+    reason: diagnostic,
+    stopReason: "ordinary_task_no_progress_guard",
+    systemMessage: diagnostic,
+  };
+}
+
 async function persistNativeStopSignature(
   stateDir: string,
   payload: CodexHookPayload,
@@ -2678,8 +2784,14 @@ async function buildStopHookOutput(
   if (ralphCompletionAuditBlock) {
     await reopenRalphCompletionAuditBlock(ralphCompletionAuditBlock);
     const blockingPath = formatStopStatePath(cwd, ralphCompletionAuditBlock.path);
-    const systemMessage =
-      `OMX Ralph completion audit is missing required evidence (${ralphCompletionAuditBlock.reason}; state: ${blockingPath}); continue verification, record a prompt-to-artifact checklist plus verification evidence, and do not report complete yet.`;
+    const systemMessage = [
+      `OMX Ralph completion audit is missing required evidence (${ralphCompletionAuditBlock.reason}; state: ${blockingPath}).`,
+      "Continue verification and do not report complete yet.",
+      "Record machine-readable completion evidence before stopping:",
+      "- either set state.completion_audit = { passed: true, prompt_to_artifact_checklist: [...], verification_evidence: [...] }",
+      "- or set completion_audit_path / completion_audit_evidence_path to a repo-relative JSON file with those same fields.",
+      "Markdown artifacts and flat top-level checklist/evidence fields are not accepted by the Ralph Stop gate.",
+    ].join(" ");
     return await returnPersistentStopBlock(
       payload,
       stateDir,
@@ -2870,6 +2982,13 @@ async function buildStopHookOutput(
         { allowRepeatDuringStopHook: true },
       );
     }
+    const ordinaryNoProgressOutput = await maybeBuildOrdinaryStopNoProgressOutput(
+      payload,
+      stateDir,
+      canonicalSessionId,
+    );
+    if (ordinaryNoProgressOutput) return ordinaryNoProgressOutput;
+
     const autoNudgeConfig = await loadAutoNudgeConfig();
     const autoNudgePhase = await readStopAutoNudgePhase(cwd, stateDir, canonicalSessionId, threadId);
 

package/src/scripts/codex-native-pre-post.ts

@@ -793,14 +793,20 @@ function buildGitCommitComplianceErrors(message: string | null): string[] {
     errors.push("Add a blank line after the subject before the narrative body.");
   }
 
+  const hasSubject = (lines[0]?.trim() ?? "") !== "";
+  const hasBlankSeparator = lines.length >= 2 && lines[1]?.trim() === "";
   const { bodyText, trailerLines } = splitBodyAndTrailerLines(lines.slice(2).join("\n"));
-  if (!bodyText) {
-    errors.push("Add a narrative body paragraph explaining the decision context.");
-  }
-  if (!trailerLines.some((line) => LORE_TRAILER_PREFIXES.some((prefix) => line.startsWith(prefix)))) {
-    errors.push("Add at least one Lore trailer such as `Constraint:`, `Confidence:`, or `Tested:`.");
+  const hasOmxCoauthorTrailer = trailerLines.includes(OMX_COAUTHOR_TRAILER);
+  const usesCompactLorePath = hasSubject && hasBlankSeparator && !bodyText && hasOmxCoauthorTrailer;
+  if (!usesCompactLorePath) {
+    if (!bodyText) {
+      errors.push("Add a narrative body paragraph explaining the decision context.");
+    }
+    if (!trailerLines.some((line) => LORE_TRAILER_PREFIXES.some((prefix) => line.startsWith(prefix)))) {
+      errors.push("Add at least one Lore trailer such as `Constraint:`, `Confidence:`, or `Tested:`.");
+    }
   }
-  if (!trailerLines.includes(OMX_COAUTHOR_TRAILER)) {
+  if (!hasOmxCoauthorTrailer) {
     errors.push(`Add the required co-author trailer: \`${OMX_COAUTHOR_TRAILER}\`.`);
   }
 

package/src/scripts/run-test-files.ts

@@ -6,6 +6,12 @@ const DEFAULT_TEST_TIMEOUT_MS = 0;
 const DEFAULT_RUNNER_TIMEOUT_MS = 30 * 60 * 1_000;
 const DEFAULT_CI_TEST_CONCURRENCY = 1;
 
+function parseBooleanEnv(value: string | undefined): boolean {
+  if (!value) return false;
+  const normalized = value.trim().toLowerCase();
+  return normalized === '1' || normalized === 'true' || normalized === 'yes' || normalized === 'on';
+}
+
 function collectTests(path: string, out: string[]): void {
   let stats;
   try {
@@ -61,6 +67,7 @@ if (files.length === 0) {
 const testTimeoutMs = parseTimeoutMs(process.env.OMX_NODE_TEST_TIMEOUT_MS, DEFAULT_TEST_TIMEOUT_MS);
 const runnerTimeoutMs = parseTimeoutMs(process.env.OMX_NODE_TEST_RUNNER_TIMEOUT_MS, DEFAULT_RUNNER_TIMEOUT_MS);
 const testConcurrency = parseTestConcurrency(process.env);
+const forceExit = parseBooleanEnv(process.env.OMX_NODE_TEST_FORCE_EXIT);
 const testArgs = ['--test'];
 if (testTimeoutMs > 0) {
   testArgs.push(`--test-timeout=${testTimeoutMs}`);
@@ -68,14 +75,17 @@ if (testTimeoutMs > 0) {
 if (testConcurrency) {
   testArgs.push(`--test-concurrency=${testConcurrency}`);
 }
+if (forceExit) {
+  testArgs.push('--test-force-exit');
+}
 testArgs.push(...files);
 
 console.error(
   `[run-test-files] running ${files.length} test file(s) from ${targets.join(', ')}${
     testTimeoutMs > 0 ? ` with per-test timeout ${testTimeoutMs}ms` : ' with per-test timeout disabled'
   }${testConcurrency ? `, test concurrency ${testConcurrency}` : ', default test concurrency'}${
-    runnerTimeoutMs > 0 ? `, and runner timeout ${runnerTimeoutMs}ms` : ', and runner timeout disabled'
-  }`,
+    forceExit ? ', force exit enabled' : ', force exit disabled'
+  }${runnerTimeoutMs > 0 ? `, and runner timeout ${runnerTimeoutMs}ms` : ', and runner timeout disabled'}`,
 );
 
 const childEnv = { ...process.env };
@@ -101,6 +111,7 @@ console.error(
     + `Roots: ${targets.join(', ')}. Test files: ${files.length}. `
     + `Per-test timeout: ${testTimeoutMs > 0 ? `${testTimeoutMs}ms` : 'disabled'}. `
     + `Test concurrency: ${testConcurrency ?? 'default'}. `
+    + `Force exit: ${forceExit ? 'enabled' : 'disabled'}. `
     + `Runner timeout: ${runnerTimeoutMs > 0 ? `${runnerTimeoutMs}ms` : 'disabled'}.`,
 );
 process.exit(1);

package/dist/planning/__tests__/context-pack-status.test.d.ts

@@ -1,2 +0,0 @@
-export {};
-//# sourceMappingURL=context-pack-status.test.d.ts.map

package/dist/planning/__tests__/context-pack-status.test.d.ts.map

@@ -1 +0,0 @@
-{"version":3,"file":"context-pack-status.test.d.ts","sourceRoot":"","sources":["../../../src/planning/__tests__/context-pack-status.test.ts"],"names":[],"mappings":""}