pair-process-plan-tasks
Breaks a refined user story into implementation tasks. Task-level idempotency: detects existing tasks and creates only missing ones. Appends condensed Technical Analysis + Task Breakdown (checklist, D
SKILL.md.claude/skills/pair-process-plan-tasks/SKILL.md
GitHub/pair-process-plan-tasks — Task Breakdown
Transform a refined user story into specific, actionable implementation tasks. Task-level idempotency: detects existing tasks in the story body and creates only missing ones. Appends an integrated Technical Analysis + Task Breakdown section to the story issue body. Tasks live inside the story — no separate task issues are created in the PM tool.
Process Profile
<!-- process-step: id=plan-tasks -->Executable form of the plan-tasks step, and a composer of define-bounded-contexts. A direct invocation while a step is disabled by the project's profile warns and asks for confirmation; a composed one never prompts — it degrades exactly as a step that is not installed. No section ⇒ no-op. See process-profile gate.
Composed Skills
| Skill | Type | Required |
|---|---|---|
/pair-capability-write-issue |
Capability | Yes — updates the story issue body with the Task Breakdown section |
Arguments
| Argument | Required | Description |
|---|---|---|
$story |
No | Story identifier (e.g., #42). If omitted, selects the highest-priority Refined story from the backlog. |
Algorithm
Step 0: Story Selection
Check: Is
$storyprovided?Skip: If provided, load the story from the PM tool and proceed to Step 1.
Act: If not provided, query the PM tool for stories in Refined state. Apply selection criteria:
- Priority: P0 > P1 > P2
- Sprint need: stories required for upcoming sprint
- Dependency chain: stories unblocking other work
Act: Present recommendation and ask developer to confirm:
Recommend task breakdown for Story
#[ID]: [Title](Priority: [P0/P1/P2]). Reason: [business value / sprint urgency / unblocks other work]. Proceed?Verify: Story selected and loaded.
Step 1: Detect Existing Tasks
Check: Read the story body. Look for a
## Task Breakdownsection with task entries (- [ ] **T-N**:or- [x] **T-N**:).Act: Classify the state:
- No Task Breakdown section → full breakdown needed (Steps 2–6).
- Partial tasks (some T-N entries but gaps in AC coverage) → resume from first missing task (Steps 3–6).
- All tasks present with full AC coverage → already complete. Offer selective update (Step 7).
Verify: Task state determined. Report:
Task state: [N existing tasks covering M/K ACs]. [Full breakdown | Resume from T-X | Already complete — offering update].
Step 2: Technical Context Analysis
Skip if: Story body already contains Technical Analysis section.
- Act: Analyze the story requirements against adoption files:
- Read architecture.md for architectural patterns.
- Read tech-stack.md for implementation choices.
- Identify affected system areas (services/modules) relevant to the story.
- Act: Present technical context summary to developer for validation.
- Verify: Developer approves. Context established.
Step 3: Task Identification
Act: Decompose the story into tasks by mapping acceptance criteria to implementation work:
- Map each AC to one or more tasks.
- Group tasks by system area (service/module) where helpful.
- Identify task types: Development, Documentation, Configuration, Research.
- Ensure every AC is covered by at least one task.
Act: Identify dependencies between tasks and determine execution order.
Act: Present task list to developer:
Proposed [N] tasks for
#[ID]: [Numbered list with task title, type, system area, estimated hours] AC coverage: [all ACs mapped] Approve or adjust?Verify: Developer approves. Task list finalized.
Step 4: Task Definition
For each task (skipping tasks that already exist in the story body):
- Act: Define the task following the task-template.md:
- Task Information (ID, priority, hours, system area)
- Summary and type
- Description with implementation context
- Acceptance criteria (deliverable, quality, integration, verification)
- Implementation approach (design, modules, files, standards)
- Dependencies (technical, task, resource)
- Implementation steps
- Testing strategy
- Verify: Each task is complete and self-contained.
Step 5: Assemble Task Breakdown Section
Act: Build the complete Task Breakdown section for the story body:
- Condensed Technical Analysis: Keep strategy summary, data flow/pipeline order, key risks. Remove verbose detail.
- Task checklist:
- [ ] **T-N**: [Title]for each task. - Dependency Graph: ASCII block showing task dependencies (see task-template.md Task Breakdown Format).
- AC Coverage table: Maps each AC to covering tasks.
- Detailed tasks: Full task definition per task (from Step 4).
Section ordering in the story body: Technical Analysis and Task Breakdown are the last content sections (after Notes, before footer).
Verify: All ACs covered. Dependency graph consistent with task dependencies. No orphan tasks.
Step 6: PM Tool Update
- Act: Compose
/pair-capability-write-issuewith$type: story,$id: [story-id]to append the Task Breakdown section to the story body. Tasks are documented inline in the story — do NOT create separate task issues. - Verify: Story body updated with Task Breakdown section.
Step 7: Already-Complete Update (optional path)
Reached only when Step 1 detects all tasks are present with full AC coverage.
Act: Ask the developer what to update:
Story
#[ID]already has a complete task breakdown. What needs updating?- Add new tasks (new AC or scope change)
- Modify existing task definitions
- Update dependency graph
- Full re-breakdown
Act: For selected option, re-execute the corresponding steps.
Act: Compose
/pair-capability-write-issuewith$type: story,$id: [story-id], and updated content.Verify: Story updated.
Output Format
TASK BREAKDOWN COMPLETE:
├── Story: [#ID: Title]
├── Tasks: [N tasks created]
├── Types: [Development: N, Documentation: N, Configuration: N]
├── AC Coverage:[K/K ACs covered]
├── PM Tool: [Story updated with Task Breakdown section]
└── Next: /pair-process-implement to start task execution
HALT Conditions
- No Refined stories in backlog (Step 0) — nothing to break down.
- Story not found or not Refined (Step 0) — must be refined first (suggest
/pair-process-refine-story). - PM tool not accessible — cannot read or update stories.
- Developer rejects task list (Step 3) — must resolve before proceeding.
- AC coverage gap (Step 5) — every AC must be covered by at least one task.
Graceful Degradation
- If
/pair-capability-write-issueis not installed, warn and provide the formatted Task Breakdown section for manual PM tool update. - If the PM tool is not accessible, produce the task breakdown content and ask the developer to update manually.
- If adoption files (architecture, tech-stack) are not found, skip technical context alignment and warn.
Notes
- This skill modifies PM tool state — it updates the story body with the Task Breakdown section. No separate task issues are created.
- Task-level idempotency: re-invoking on a story with partial tasks appends only missing ones to the story body.
- The integrated Technical Analysis + Task Breakdown format positions these as the last content sections in the story body (functional sections first, technical sections last).
- Condensed TA retains: strategy summary, data flow/pipeline order, key risks. Removes verbose implementation detail that belongs in individual tasks.
- Task template reference: task-template.md.
- Story template reference: user-story-template.md.