القائمة
Convert an execution-ready prompt into a concrete task backlog with sequence, ownership, dependencies, and validation steps. Use after prompt generation and before orchestration.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Generate clear, ready-to-paste prompts for Codex that turn rough requests into executable instructions with scope, constraints, acceptance criteria, and validation steps. Use when you want a high-quality prompt before asking Codex to implement features, fix bugs, refactor code, review changes, write tests, or perform deployments.
Execute implementation work in the Dolphi codebase using the project's policy, workflow, and response template. Use when handling any feature, bug fix, refactor, test, or documentation request in this repository that must comply with offline-first privacy, BRD/PRD scope, and change-control rules.
Maintain and synchronize Dolphi repository documentation after structural, policy, architecture, testing, or workflow changes. Use when any update may impact context.md, agent.md, roadmap, architecture, standards, DB spec, module docs, or testing policy files.
Publish local repository changes to GitHub with a predictable non-interactive flow. Use when the user asks to push local changes to a target GitHub remote and branch.
Create or update repository-local skills required by the task plan. Use when the orchestrator detects missing reusable capabilities.
Orchestrate execution using a strict chain of prompt generation, task planning, optional skill generation, and implementation handoff. Always enforce plan-first behavior.
| name | task-generator |
| description | Convert an execution-ready prompt into a concrete task backlog with sequence, ownership, dependencies, and validation steps. Use after prompt generation and before orchestration. |
Transform a prompt into an actionable task list that an orchestrator can execute safely.
$codex-prompt-generatordocs/architecture/KARPATHY_CODING_GUIDELINES.md.Return only this structure:
Task Plan:
1. <task title>
- Owner: <role or skill>
- Inputs: <required inputs>
- Outputs: <expected artifacts>
- Dependencies: <task ids or none>
- Validation: <commands/checks>
- Risk: <low|medium|high + note>
Execution Order:
1. <ordered sequence with optional parallel groups>
Blockers:
- <blocker or none>
Missing Required Inputs:
- <missing input list or none>