Agent skill
implement-plan
Implement a feature plan from a spec file, following test-first methodology and tracking all deliverables
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/implement-plan-incubateur-ademe-benefriches
SKILL.md
Implement Plan
Follow the phases below to implement the Plan, then Report the completed work.
IMPORTANT: If no plan is provided (i.e., $ARGUMENTS is empty or blank), STOP immediately and ask the user to provide one. Example usage: /implement-plan specs/new-feature.md
Plan
$ARGUMENTS
Phase 1: Parse Deliverables
Before writing any code, extract ALL deliverables from the plan:
-
Read the ENTIRE plan from start to finish
-
Find the "Files to Deliver" section (or extract from "Step by Step Tasks" if not present)
-
Detect database work: If the plan mentions any of these, use
/create-database-migrationskill:- Database migrations (new tables, columns, schema changes)
- Updates to
tableTypes.d.ts - SQL schema modifications
- Data migrations
-
Create a structured inventory using TodoWrite:
Test Files (will be implemented first):
- Create a todo for each test file with format: "Create test:
path/to/file.spec.ts" - Under each test file, note the test cases it must contain
Production Files (will be implemented after tests):
- Create a todo for each new file: "Create:
path/to/file.ts" - Create a todo for each modification: "Modify:
path/to/file.ts- "
Validation:
- Create a todo for each validation command from the plan
- Create a todo for each test file with format: "Create test:
-
Review the inventory: Does it capture EVERYTHING from the plan? If not, add missing items.
Phase 2: Implement Tests First
For each test file in your inventory:
- Mark the test todo as
in_progress - Create the test file with ALL test cases described in the plan
- Tests should initially fail (production code doesn't exist yet) - this is expected
- Mark the test todo as
completed
WHY TESTS FIRST:
- Tests act as a contract that forces production code to exist
- Forgotten tests = forgotten functionality (caught immediately)
- Provides deterministic verification that all production code is generated
Phase 3: Implement Production Code
For each production file in your inventory:
- Mark the file todo as
in_progress - Implement the code following the plan's specifications
- Run related tests to verify the implementation
- Tests should now pass - if they don't, fix the implementation
- Mark the file todo as
completed
Continue until all production todos are complete AND all tests pass.
Phase 4: Final Verification
Before reporting:
- Re-read the plan's "Files to Deliver" or "Step by Step Tasks" section
- Cross-check: Does every listed file exist?
- Cross-check: Does every listed test case exist and pass?
- Run ALL validation commands from the plan
- If anything is missing, go back and complete it
Report
Only after Phase 4 is complete:
- Summarize the work in a concise bullet point list
- Show files and lines changed:
git diff --stat - Confirm: "All tests pass" and "All validation commands pass"
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?