Intent Integrity Kit Tasks

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

1. Run: bash .tessl/tiles/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/bash/check-prerequisites.sh --phase 05 --json Windows: pwsh .tessl/tiles/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/powershell/check-prerequisites.ps1 -Phase 05 -Json

2. 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.

3. 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/tiles/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/bash/set-active-feature.sh --json <selection>

   Windows: pwsh .tessl/tiles/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/powershell/set-active-feature.ps1 -Json <selection>

   Then re-run the prerequisites check from step 1.

4. Checklist gate per checklist-gate.md.

Plan Readiness Validation

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

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

Execution Flow

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").

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.

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.

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 — missing required elements:

• - [ ] Create User model (no ID, no story label)
• T001 [US1] Create model (no checkbox)
• - [ ] [US1] Create User model (no task ID)

Traceability: When referencing multiple test spec IDs, enumerate them explicitly as a comma-separated list. Do NOT use English prose ranges like "TS-005 through TS-010" — these break automated traceability checks.

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

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

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.

7. Dependency Graph Validation

After generating, validate:

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

8. Write tasks.md

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

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

git add specs/*/tasks.md
git commit -m "tasks: <feature-short-name> task breakdown"

Dashboard Refresh

Regenerate the dashboard so the pipeline reflects the new tasks:

bash .tessl/tiles/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/bash/generate-dashboard-safe.sh

Windows: pwsh .tessl/tiles/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/powershell/generate-dashboard-safe.ps1

Next Steps

Run: bash .tessl/tiles/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/bash/next-step.sh --phase 05 --json Windows: pwsh .tessl/tiles/tessl-labs/intent-integrity-kit/skills/iikit-core/scripts/powershell/next-step.ps1 -Phase 05 -Json

Parse the JSON and present:

1. If clear_after is true: suggest /clear before proceeding
2. Present next_step as the primary recommendation
3. If alt_steps non-empty: list as alternatives
4. For next_step and each alt_step, include the model_tier from the JSON so the user knows which model is best for each option. Look up tiers in model-recommendations.md for agent-specific switch commands.
5. Append dashboard link

Format:

Tasks generated!
Next: [/clear → ] <next_step> (model: <tier>)
[- <alt_step> — <reason> (model: <tier>)]

- Dashboard: file://$(pwd)/.specify/dashboard.html (resolve the path)