codex-workflows 0.6.5 → 0.6.7
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/.agents/skills/documentation-criteria/references/plan-template.md +27 -0
- package/.agents/skills/documentation-criteria/references/task-template.md +12 -0
- package/.agents/skills/recipe-build/SKILL.md +20 -6
- package/.agents/skills/recipe-front-build/SKILL.md +20 -6
- package/.agents/skills/recipe-front-plan/SKILL.md +12 -0
- package/.agents/skills/recipe-front-review/SKILL.md +2 -1
- package/.agents/skills/recipe-fullstack-build/SKILL.md +20 -6
- package/.agents/skills/recipe-implement/SKILL.md +1 -1
- package/.agents/skills/recipe-plan/SKILL.md +14 -1
- package/.agents/skills/recipe-review/SKILL.md +2 -1
- package/.agents/skills/subagents-orchestration-guide/SKILL.md +14 -3
- package/.agents/skills/subagents-orchestration-guide/references/monorepo-flow.md +12 -4
- package/.codex/agents/acceptance-test-generator.toml +25 -5
- package/.codex/agents/code-reviewer.toml +5 -1
- package/.codex/agents/document-reviewer.toml +18 -1
- package/.codex/agents/integration-test-reviewer.toml +14 -2
- package/.codex/agents/task-decomposer.toml +16 -0
- package/.codex/agents/task-executor-frontend.toml +12 -15
- package/.codex/agents/task-executor.toml +12 -15
- package/.codex/agents/technical-designer-frontend.toml +10 -81
- package/.codex/agents/technical-designer.toml +10 -94
- package/.codex/agents/work-planner.toml +28 -1
- package/package.json +1 -1
|
@@ -46,6 +46,8 @@ Skill Status:
|
|
|
46
46
|
8. Propagate UI Spec component and runtime boundary context into the plan so task decomposition can pass it to executors
|
|
47
47
|
9. Map implementation-binding ADR decisions to constrained tasks
|
|
48
48
|
10. Document in progress-trackable format
|
|
49
|
+
11. Carry proof obligations from test skeletons or acceptance criteria into the work plan so task files and reviews can judge whether tests prove the claim, not merely run
|
|
50
|
+
12. Record WorkPlan Review status as pending when creating or materially updating a plan
|
|
49
51
|
|
|
50
52
|
## Input Parameters
|
|
51
53
|
|
|
@@ -70,6 +72,7 @@ Read the Design Doc(s), UI Spec, PRD, and ADR (if provided). Extract:
|
|
|
70
72
|
- Quality Assurance Mechanisms from each Design Doc: all items marked `adopted`, including mechanism name, enforced quality aspect, configuration path, and covered files or project-wide scope
|
|
71
73
|
- Implementation-relevant technical requirements from each Design Doc section using the category values defined in the plan template's `Design-to-Plan Traceability` section
|
|
72
74
|
- Prerequisite ADR references from each Design Doc and implementation-binding decisions from each resolved ADR
|
|
75
|
+
- Primary failure modes from acceptance criteria or generated test skeletons
|
|
73
76
|
|
|
74
77
|
Focus on implementation-relevant items only: items that directly inform task creation, dependency ordering, verification design, or protected no-change boundaries.
|
|
75
78
|
Extract `scope-boundary` rows from explicit non-target statements such as `No Ripple Effect`, compatibility constraints, or protected boundaries that must remain unchanged.
|
|
@@ -88,8 +91,11 @@ Choose Strategy A (TDD) if test skeletons are provided, Strategy B (implementati
|
|
|
88
91
|
### 4. Compose Phases
|
|
89
92
|
|
|
90
93
|
**Common rules (all approaches)**:
|
|
94
|
+
- Record Review Scope in the plan header as the planned file scope derived from the Design Doc and task targets. For revision plans, record the base branch and diff range.
|
|
95
|
+
- Include a `WorkPlan Review` section. Set `Status: pending` and `Conditions: none`. When invoked in update mode after a WorkPlan review verdict, only `approved` with no conditions may set `Status: approved`.
|
|
91
96
|
- Preserve Verification Strategies per Design Doc in the work plan header and keep each source document path. Merge strategies only when the Design Docs explicitly define a shared one
|
|
92
97
|
- Include Verification Strategy summaries in the work plan header so the plan is self-sufficient for downstream task generation
|
|
98
|
+
- Include a Proof Strategy in the work plan header. When test skeletons exist, use their Primary failure mode and Proof obligation annotations as the proof obligation source; otherwise derive each claim's primary failure mode from the acceptance criterion and Design Doc verification strategy.
|
|
93
99
|
- Include adopted Quality Assurance Mechanisms in the work plan header so downstream task generation and quality checks inherit the intended quality gates
|
|
94
100
|
- Place tasks with the lowest dependencies in earlier phases
|
|
95
101
|
- Map normalized verification timing to phases as follows: `phase_1` -> earliest implementation phase, `per_phase` -> each relevant phase, `integration_phase` -> integration phase, `final_phase` -> final Quality Assurance phase
|
|
@@ -171,8 +177,22 @@ Mapping rules:
|
|
|
171
177
|
- Acceptance criteria and required user-visible behaviors belong in `Design-to-Plan Traceability`; `ADR Bindings` covers structural implementation constraints.
|
|
172
178
|
- If an ADR decision constrains the design but no task covers it, add a justified gap and flag it for user confirmation before plan approval.
|
|
173
179
|
|
|
180
|
+
### 5d. Build Failure Mode Checklist
|
|
181
|
+
|
|
182
|
+
Populate the plan template's `Failure Mode Checklist` before finalizing tasks. Enumerate all eight domain-independent categories, mark whether each applies, and list the task IDs that cover applicable categories. Keep category names generic and put project-specific details in task descriptions or notes.
|
|
183
|
+
|
|
184
|
+
Categories:
|
|
185
|
+
- same-value
|
|
186
|
+
- no-op
|
|
187
|
+
- empty input
|
|
188
|
+
- invalid option
|
|
189
|
+
- missing config
|
|
190
|
+
- unavailable boundary
|
|
191
|
+
- shared-state dependency
|
|
192
|
+
- rollback-only visibility
|
|
193
|
+
|
|
174
194
|
### 6. Define Tasks with Completion Criteria
|
|
175
|
-
For each task, derive completion criteria from Design Doc acceptance criteria. Apply the 3-element completion definition (Implementation Complete, Quality Complete, Integration Complete).
|
|
195
|
+
For each task, derive completion criteria from Design Doc acceptance criteria. Each task that covers an acceptance criterion must explicitly name the AC ID or claim identifier in Completion Criteria or Proof Obligations. Apply the 3-element completion definition (Implementation Complete, Quality Complete, Integration Complete).
|
|
176
196
|
|
|
177
197
|
### 7. Produce Work Plan Document
|
|
178
198
|
Write the work plan following the plan template from documentation-criteria skill. Include Phase Structure Diagram and Task Dependency Diagram (mermaid).
|
|
@@ -244,6 +264,8 @@ Read available test skeleton files (integration tests, and E2E tests only when p
|
|
|
244
264
|
- `// @dependency:` → Dependent components (material for phase placement decisions)
|
|
245
265
|
- `// @complexity:` → Complexity (high/medium/low, material for effort estimation)
|
|
246
266
|
- `// Value Score:` → Priority judgment
|
|
267
|
+
- `// Primary failure mode:` → Regression the implemented test must detect
|
|
268
|
+
- `// Proof obligation:` → Boundary, state, and mock-rationale obligations the implemented test must satisfy
|
|
247
269
|
|
|
248
270
|
#### Step 2: Reflect Meta Information in Work Plan
|
|
249
271
|
|
|
@@ -257,6 +279,11 @@ Read available test skeleton files (integration tests, and E2E tests only when p
|
|
|
257
279
|
- `// @complexity: high` → Subdivide tasks or estimate higher effort
|
|
258
280
|
- `// @complexity: low` → Consider combining multiple tests into one task
|
|
259
281
|
|
|
282
|
+
3. **Proof Strategy Propagation**
|
|
283
|
+
- Copy Primary failure mode and Proof obligation annotations into the plan's Proof Strategy
|
|
284
|
+
- Map each proof obligation to the task that implements or verifies the corresponding acceptance criterion or user journey
|
|
285
|
+
- When a proof obligation cannot be fully satisfied until a later phase, record the residual and the closing task
|
|
286
|
+
|
|
260
287
|
#### Step 3: Extract Environment Prerequisites from E2E Skeletons
|
|
261
288
|
|
|
262
289
|
When E2E test skeletons are provided, first identify the E2E skeleton subset using file naming conventions such as `*.fixture.e2e.test.*`, `*.service.e2e.test.*`, or `*.e2e.test.*`, then scan only those files for environment prerequisites in two stages:
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "codex-workflows",
|
|
3
|
-
"version": "0.6.
|
|
3
|
+
"version": "0.6.7",
|
|
4
4
|
"description": "Task-oriented agentic coding framework for OpenAI Codex CLI — skills, recipes, and subagents for structured development workflows",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"author": "Shinsuke Kagawa",
|