iikit-05-tasks - SKILL.md Agent Skill

name: iikit-05-tasks description: >- Generate dependency-ordered task breakdown from plan and specification. Use when breaking features into implementable tasks, planning sprints, or creating work items with parallel markers. license: MIT metadata: version: "1.6.4"

Intent Integrity Kit Tasks

Process steps in order. Do not skip ahead.

Generate an actionable, dependency-ordered tasks.md for the feature.

User Input

$ARGUMENTS

You MUST consider the user input before proceeding (if not empty).

Constitution Loading

Load constitution per constitution-loading.md (basic mode — note TDD requirements for task ordering).

Prerequisites Check

Run: bash .tessl/plugins/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/bash/check-prerequisites.sh --phase 05 --json Windows: pwsh .tessl/plugins/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/powershell/check-prerequisites.ps1 -Phase 05 -Json
Parse JSON for FEATURE_DIR and AVAILABLE_DOCS. If missing plan.md: ERROR. If script exits with testify error: STOP and tell the user to run /iikit-04-testify first.
If JSON contains needs_selection: true: present the features array as a numbered table (name and stage columns). Follow the options presentation pattern in conversation-guide.md. After user selects, run:
```
bash .tessl/plugins/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/bash/set-active-feature.sh --json <selection>
```
Windows: pwsh .tessl/plugins/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/powershell/set-active-feature.ps1 -Json <selection>

Then re-run the prerequisites check (item 1 of this section).
Checklist gate per checklist-gate.md.

Plan Readiness Validation

Tech stack: verify plan.md has Language/Version defined (WARNING if missing)
User story mapping: verify each story in spec.md has acceptance criteria
Dependency pre-analysis: identify shared entities used by multiple stories -> suggest Foundational phase

Report readiness per formatting-guide.md (Plan Readiness section).

Step 1 — Load Documents

Required: plan.md, spec.md
Optional: data-model.md, contracts/, research.md, quickstart.md, tests/features/ (.feature files)

If .feature files exist (or legacy test-specs.md), tasks reference specific test IDs (e.g., "T012 [US1] Implement to pass TS-001").

Proceed immediately to Step 2.

Step 2 — Tessl Convention Consultation

If Tessl installed: query primary framework tile for project structure conventions and testing framework tile for test organization. Apply to file paths and task ordering. If not available: skip silently.

Proceed immediately to Step 3.

Step 3 — Generate Tasks

Extract tech stack from plan.md, user stories from spec.md, entities from data-model.md, endpoints from contracts/, decisions from research.md. Organize by user story with dependency graph and parallel markers.

Proceed immediately to Step 4.

Step 4 — Task Format (REQUIRED)

- [ ] [TaskID] [P?] [Story?] Description with file path

Checkbox: always - [ ]
Task ID: sequential (T001, T002...)
[P]: only if parallelizable (different files, no dependencies)
[USn]: required for user story tasks only (not Setup/Foundational/Polish)
Description: clear action with exact file path

Examples:

- [ ] T001 Create project structure per implementation plan (setup, no story)
- [ ] T005 [P] Implement authentication middleware in src/middleware/auth.py (parallel, no story)
- [ ] T012 [P] [US1] Create User model in src/models/user.py (parallel + story)
- [ ] T014 [US1] Implement UserService in src/services/user_service.py (sequential + story)

Wrong — each example below is missing a required element:

No checkbox: T001 [US1] Create model
No task ID: - [ ] [US1] Create User model
No story label (on a story task): - [ ] T014 Create UserService

Traceability: Enumerate multiple test spec IDs explicitly as a comma-separated list — do NOT use prose ranges (e.g., "TS-005 through TS-010"), which break automated traceability checks.

Correct: [TS-005, TS-006, TS-007, TS-008, TS-009, TS-010] Wrong: TS-005 through TS-010

Proceed immediately to Step 5.

Step 5 — Phase Structure

Phase 1: Setup (project initialization)
Phase 2: Foundational (blocking prerequisites, complete before stories)
Phase 3+: User Stories in priority order (P1, P2, P3...) — tests -> models -> services -> endpoints -> integration
Final: Polish & Cross-Cutting Concerns

Proceed immediately to Step 6.

Step 6 — Task Organization

Map each component to its user story. Shared entities serving multiple stories go in Setup/Foundational. Each contract gets a contract test task. Story dependencies marked explicitly.

Proceed immediately to Step 7.

Step 7 — Dependency Graph Validation

After generating, validate:

Circular dependencies: detect cycles, ERROR if found with resolution options
Orphan tasks: warn about tasks with no dependencies and not blocking anything
Critical path: identify longest chain, suggest parallelization, list parallel batches per phase
Phase boundaries: no backward cross-phase dependencies
Story independence: warn on priority inversions (higher-priority depending on lower)

Proceed immediately to Step 8.

Step 8 — Write tasks.md

Use tasks-template.md with phases, dependencies, parallel examples, and implementation strategy.

Finish here.

Report

Output: path to tasks.md, total count, count per story, parallel opportunities, MVP scope suggestion, format validation.

Semantic Diff on Re-run

If tasks.md exists: preserve [x] completion status, map old IDs to new by similarity, warn about changes to completed tasks. Ask confirmation before overwriting. Use format from formatting-guide.md (Semantic Diff section).

Commit, Dashboard & Next Steps

Run post-phase to commit, refresh dashboard, and compute next step in a single call:

bash .tessl/plugins/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/bash/post-phase.sh --phase 05 --commit-files "specs/*/tasks.md" --commit-msg "tasks: <feature-short-name> task breakdown"

Windows: pwsh .tessl/plugins/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/powershell/post-phase.ps1 -Phase 05 -CommitFiles "specs/*/tasks.md" -CommitMsg "tasks: <feature-short-name> task breakdown"

Parse next_step from JSON. Present per model-recommendations.md:

Tasks generated!
Next: [/clear → ] <next_step> (model: <tier>)
[- <alt_step> — <reason> (model: <tier>)]
- Dashboard: file://$(pwd)/.specify/dashboard.html