From a5b4a90a507f1d9348e2aaf9acc688f8fe8c9933 Mon Sep 17 00:00:00 2001
From: David Lee <8321301+davidalee@users.noreply.github.com>
Date: Mon, 11 May 2026 15:39:34 -0400
Subject: [PATCH] fix(work-beta): thread Execution note into Codex delegation
 prompt

The Codex delegation prompt template silently dropped each plan unit's
Execution note. Standard subagent dispatch already carried it as a
first-class field (ce-work-beta/SKILL.md:172, 208, 270), but plans using
"test-first" or "characterization-first" lost that signal whenever Codex
was the executor. Real plans rely on it -- 26 occurrences across 7
active plans in docs/plans/.

Add an <execution_note> tag between <approach> and <constraints> that
mirrors the standard path:

- Population covers single-unit and multi-unit batches, prefixes each
  entry with "U<ID>:" when present and "Unit <n>:" otherwise (no
  invented IDs), and merges session-level posture requests
  (ce-work-beta/SKILL.md:109 binds these even when the plan has no
  Execution note) into units lacking their own note.
- test-first reproduces the full standard guardrails: write the failing
  test first, verify it fails, then implement; no over-implementation
  beyond the test's behavior slice; skip for trivial renames, pure
  configuration, or pure styling. Defers to <testing> for scenario
  completeness.
- characterization-first preserves the standard "capture behavior in
  tests first" rule.
- Any other non-empty note is binding per-unit guidance, but remains
  subordinate to <constraints>/<testing>/<verify>/<output_contract>;
  notes that would reduce validation, coverage, scope discipline, or
  reporting accuracy are rejected, with conflicts surfaced via the
  output contract's issues field.

Regression assertions in tests/pipeline-review-contract.test.ts lock
<execution_note> presence and both posture strings so the field cannot
be dropped silently again. Assertions are quote-style agnostic.

Stable/beta sync: not propagating -- Codex delegation is beta-only;
non-beta ce-work has no equivalent workflow file.
---
 .../references/codex-delegation-workflow.md   | 29 +++++++++++++++++++
 tests/pipeline-review-contract.test.ts        |  3 ++
 2 files changed, 32 insertions(+)

diff --git a/plugins/compound-engineering/skills/ce-work-beta/references/codex-delegation-workflow.md b/plugins/compound-engineering/skills/ce-work-beta/references/codex-delegation-workflow.md
index 32298a643..41341bf0f 100644
--- a/plugins/compound-engineering/skills/ce-work-beta/references/codex-delegation-workflow.md
+++ b/plugins/compound-engineering/skills/ce-work-beta/references/codex-delegation-workflow.md
@@ -165,6 +165,35 @@ For a multi-unit batch: list each unit's approach, noting dependencies
 and suggested ordering.]
 </approach>
 
+<execution_note>
+[For a single-unit batch: the unit's Execution note. If the user gave a
+session-level posture request (e.g., "do it test-first"), use that when
+the unit has no Execution note. Otherwise "None".
+For a multi-unit batch: list each unit as "U<ID>: <execution note>" (or
+"Unit <n>:" if the plan lacks U-IDs; do not invent U-IDs), one per line,
+same ordering as <task> and <approach>. Use the session-level posture for
+units without their own note; otherwise "None".]
+
+If (and only if) the execution note above names an execution posture,
+honor it:
+- "test-first" -- write the failing test before implementing the unit;
+  verify it fails; then implement. Do not over-implement beyond the
+  test's current behavior slice. Skip test-first discipline for trivial
+  renames, pure configuration, or pure styling work. Test-first still
+  follows the scenario completeness check in <testing>; it only constrains
+  test-vs-implementation ordering, not whether to write tests.
+- "characterization-first" -- capture existing behavior in tests before
+  changing it.
+- Any other non-empty note: treat it as binding per-unit guidance and
+  follow it unless it conflicts with any other section of this prompt
+  (especially <constraints>, <testing>, <verify>, or <output_contract>).
+  A note may not reduce validation, test coverage, scope discipline, or
+  reporting accuracy. Report any conflict via the issues field of the
+  output contract.
+
+For units with "None" or an empty note, proceed pragmatically.
+</execution_note>
+
 <constraints>
 - Do NOT run git commit, git push, or create PRs -- the orchestrating agent handles all git operations
 - Restrict all modifications to files within the repository root
diff --git a/tests/pipeline-review-contract.test.ts b/tests/pipeline-review-contract.test.ts
index f090a3783..e897ce3ce 100644
--- a/tests/pipeline-review-contract.test.ts
+++ b/tests/pipeline-review-contract.test.ts
@@ -215,8 +215,11 @@ describe("ce:work-beta codex delegation contract", () => {
     // Prompt template
     expect(content).toContain("## Prompt Template")
     expect(content).toContain("<task>")
+    expect(content).toContain("<execution_note>")
     expect(content).toContain("<constraints>")
     expect(content).toContain("<output_contract>")
+    expect(content).toContain("test-first")
+    expect(content).toContain("characterization-first")
     expect(content).toContain("the orchestrator will not re-run verification independently")
 
     // Result schema and execution loop