Agent skill
implementer
Execute feature tasks one by one with live documentation and context logging. Triggers on "Execute FEAT-XXX tasks", "Start implementing FEAT-XXX", "Work on FEAT-XXX", "Continue FEAT-XXX".
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/implementer
SKILL.md
Implementer
Execute feature tasks one by one with live documentation and context updates.
Triggers
- "Execute FEAT-XXX tasks"
- "Start implementing FEAT-XXX"
- "Work on FEAT-XXX"
- "Continue FEAT-XXX"
- "Resume FEAT-XXX"
- "Next task for FEAT-XXX"
Prerequisites
- Feature folder exists with completed:
spec.mdanalysis.mddesign.mdtasks.mdcontext/folder with templates
- Git branch created for feature
- If missing: guide user to complete previous phases
Purpose
- Find next uncompleted task in tasks.md
- Execute task (write code, create files, etc.)
- Update documentation in real-time
- Update context (session_log, decisions, blockers)
- Commit after each task
- Repeat until all tasks done or blocker hit
Process
1. Read Context
# Read tasks
cat docs/features/FEAT-XXX/tasks.md
# Read design for reference
cat docs/features/FEAT-XXX/design.md
# Read spec for requirements
cat docs/features/FEAT-XXX/spec.md
# Check current status
cat docs/features/FEAT-XXX/status.md
# Read last session log entries
tail -30 docs/features/FEAT-XXX/context/session_log.md
# Check for active blockers
cat docs/features/FEAT-XXX/context/blockers.md
# Verify git branch
git branch --show-current
2. Verify Branch
If on main/master: -> STOP and say: "Create feature branch first: git checkout -b feature/XXX-name"
If on correct feature branch: -> Continue with implementation
3. Find Next Task
Scan tasks.md for first task that is:
[ ]Pending, OR[in progress]In Progress (resume)
Skip tasks that are:
[x]Complete[skipped]Skipped[blocked]Blocked
4. Execute Task Loop
For each task:
A. BEFORE starting:
- Update tasks.md: [ ] -> [in progress]
- Announce: "Starting task: [task description]"
B. DURING execution:
- Write code / create files
- Follow design.md specifications
- If unclear -> ask user, don't assume
- If blocked -> mark [blocked], log to blockers.md, move to next
- If decision needed -> log to decisions.md
C. AFTER completing:
- Update tasks.md: [in progress] -> [x]
- Update progress table in tasks.md
- Commit: git add . && git commit -m "FEAT-XXX: [task]"
- Announce: "Task complete. Progress: X/Y tasks"
D. CHECKPOINT (every 30 min or 3 tasks):
- Update status.md with progress
- Update context/session_log.md
- git push origin [branch]
- Summary of progress
5. Commit Format
git add [specific files for this task]
git commit -m "FEAT-XXX: [Task description]"
Examples:
FEAT-001: Create User model
FEAT-001: Add user validation schemas
FEAT-001: Implement UserService CRUD methods
FEAT-001: Add POST /users endpoint
FEAT-001: Add unit tests for UserService
6. Completion
When all tasks done:
FEAT-XXX implementation complete!
Summary:
- Total tasks: X
- Completed: X
- Skipped: X
- Blocked: X
Next steps:
1. Push: git push -u origin feature/XXX-name
2. Create PR: /git pr
3. After merge: /wrap-up FEAT-XXX
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?