home/productivity/agent-task-4

Agent Task

GPTClaudeDeepSeek··892 copies·updated 2026-07-14
agent-task-4.prompt
# GitHub Copilot Agent Task Template

## Target Tool

GitHub Copilot coding agent (assigned to a GitHub issue, producing a draft
pull request) or Copilot Chat/agent mode inside VS Code. This template also
works as the basis for a `.github/copilot-instructions.md`-style repository
convention: the same scope, boundary, and checklist language can be copied
into that file so Copilot picks it up automatically for every session in
this repository.

## Purpose

Use this template for a small issue assignment that should produce a
reviewable branch or pull request, or for a bounded Copilot Chat agent-mode
session in the editor. It keeps the task small enough that a human can
review the resulting diff without reading the full agent transcript.

## Inputs To Fill

| Input | Description | Example |
| --- | --- | --- |
| `{issue_title}` | Short, specific issue title. | `Expand public repo safety checklist` |
| `{files_expected}` | Files the change should touch. | `docs/workflows/public-repo-safety.md` |
| `{acceptance_criteria}` | Concrete, checkable criteria. | `Includes secret scan step, working links, and CI log reference` |
| `{checks}` | GitHub Actions and/or local commands to require. | `GitHub Actions and local unittest/repo-health checks` |
| `{reviewer}` | Who reviews and merges the PR. | `Repository maintainer, squash merge only` |

## Full Prompt / Issue Body

fill the variables

This prompt has 5 variables. Pro fills them into a ready-to-paste prompt for you — no manual find-and-replace.

{issue_title}{files_expected}{acceptance_criteria}{checks}{reviewer}
Unlock with Pro →

when to use it

Community prompt sourced from the open-source GitHub repo Yaked1/ai-lab-codex-workbench (MIT). A "Agent Task" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.

tags

productivitycommunitydeveloper

source

Yaked1/ai-lab-codex-workbench · MIT