aspiers/ai-config / .agents/skills/task-implementation
task-implementation
VerifiedImplements one numbered sub-task from a task list at .ai/[feature]/tasks.md, running linters and tests, without staging, committing, or marking the task complete. Use when asked to carry out a specific sub-task by number, or as the inner step of a larger loop that handles commits separately. For the full loop including commits, use task-orchestration.
Task Implementation
Implement a single sub-task from a task list while adhering to project quality controls.
When to Use This Skill
Use this skill when:
- Working on feature development with existing task lists in
.ai/[feature_name]/tasks.md - Implementing a specific sub-task assigned to you
- Following a structured development workflow
Input
This skill expects:
feature_name- Locates task list at.ai/[feature_name]/tasks.mdsubtask_number- The specific sub-task number to implement
If not provided, ask the user to clarify.
Approach
You are a junior engineer implementing a sub-task. Do your best to ensure the implementation adheres to the project's quality controls. Run linters and tests appropriately during iterative development.
Prohibited Actions
IMPORTANT! You must NOT attempt to commit or even stage your changes in git, as that will be handled elsewhere.
While you can update details in the task list as you work, you must NOT mark tasks as completed. It is not your responsibility to judge whether a task is completed.
Process
Follow these steps EXACTLY! NO EXCEPTIONS!
-
Implement the sub-task according to your best judgment
-
Add test coverage according to repository guidelines
-
Run linters according to repository guidelines:
- First look for linting commands in:
CLAUDE.md,.cursorrules,AGENTS.md,AGENT.md,GEMINI.md - Then repository documentation (
README.md,docs/) - Then package configuration (
package.json,Makefile, etc.) - Then standard linter patterns
For each linter found: a. Run auto-fix mode if available (e.g.,
prettier,eslint --fix) b. Run check mode to see remaining issues c. If issues can't be fixed, stop and ask the user what to do nextIf not passing, go back to step 1.
- First look for linting commands in:
-
Run tests according to repository guidelines:
- Look for test commands in documentation, package config, or standard patterns
- For each test command found: run it, fix issues if possible, ask user if not
If not passing, go back to step 1.
Context Maintenance
- Update details in the task list as you work
- Do NOT mark tasks as completed
- Add new tasks if they emerge during implementation
- Update the corresponding
prp.txtif appropriate - Maintain the "Relevant Files" section:
- List every file created or modified
- Give each file a one-line description of its purpose
Quality Gates
Before considering the sub-task complete:
- Implementation matches the sub-task description
- Linting passes (auto-fixable issues resolved)
- Tests pass (100% success)
- Code follows project conventions
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 318k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 318k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 318k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |