Menu
Set up and manage GitHub Actions workflows that use Copilot coding agents for automated PR handling and issue resolution
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
| name | agentic-workflows |
| description | Set up and manage GitHub Actions workflows that use Copilot coding agents for automated PR handling and issue resolution |
| compatibility | >=3.2 |
Skill metadata: version "1.0"; license MIT; tags [github-actions, automation, ci, agents, pull-requests]; compatibility ">=3.2"; recommended tools [codebase, editFiles, runCommands, githubRepo].
Set up GitHub Actions workflows that invoke Copilot coding agents to automate issue resolution, PR creation, and code review via GitHub events.
copilot-setup-steps.yml exists (provides agent environment)contents: write, pull-requests: write, issues: read)Headless Copilot session triggered by a workflow. Reads issue/event context, plans a fix, implements it, and opens a PR using a Codex-class model.
Reusable workflow configuring the agent's environment (runtime, deps, tools). Runs before the agent. Template at template/copilot-setup-steps.yml.
| Trigger | Use case |
|---|---|
issues.labeled | Auto-assign agent when a specific label is added |
issue_comment.created | Agent responds to a command comment (e.g., /fix) |
pull_request.opened | Auto-review new PRs |
workflow_dispatch | Manual agent invocation |
Verify setup steps exist — Check that .github/workflows/copilot-setup-steps.yml exists. If not, fetch and write it:
https://raw.githubusercontent.com/asafelobotomy/copilot-instructions-template/main/template/copilot-setup-steps.yml
Choose the trigger pattern — Ask the user which event should invoke the agent:
| Pattern | Workflow trigger | Agent receives |
|---|---|---|
| Label-based | issues.labeled | Issue body and comments |
| Comment-based | issue_comment.created | Comment text and issue context |
| PR review | pull_request.opened | PR diff and description |
| Manual | workflow_dispatch | User-provided inputs |
Create the workflow — Write .github/workflows/copilot-agent.yml:
name: Copilot Agent
on:
issues:
types: [labeled]
jobs:
agent:
if: github.event.label.name == 'copilot'
runs-on: ubuntu-latest
permissions:
contents: write
pull-requests: write
issues: read
steps:
- uses: actions/checkout@v4
- uses: ./.github/workflows/copilot-setup-steps.yml
# The Copilot coding agent handles the rest automatically
Adjust the trigger and condition based on the user's choice in step 2.
Configure guardrails — Agent inherits .github/copilot-instructions.md automatically. Add label filter, least-privilege permissions, and optional concurrency group.
Test — Create a test issue, apply trigger label/comment, verify in Actions tab.
Document — Add trigger phrase and label to AGENTS.md.
permissions: write-all — scope exactly| Risk | Waste code | Mitigation |
|---|---|---|
| Agent runs on irrelevant issues | W1 Overproduction | Use specific labels and filters |
| Agent waits for human review | W2 Waiting | Enable auto-merge for low-risk changes only |
| Duplicate agent runs | W5 Inventory | Add concurrency groups |
| Over-trusting agent output | W16 Over-trust | Require human review on all agent PRs |
Health check procedures D1–D14 for the Audit agent — structural validation, attention budget, version checks, workspace integrity, and static audit
Configure and manage Model Context Protocol servers for external tool access
Review a UI for accessibility — WCAG 2.1 AA compliance, semantic HTML, ARIA usage, keyboard navigation, focus management, colour contrast, and screen reader compatibility
Design or review a REST or GraphQL API — resource modeling, versioning strategy, error contract, OpenAPI/schema-first workflow, and security baseline
Generate a CHANGELOG.md entry from staged changes, a commit range, or a PR diff — following Keep a Changelog format with conventional commit classification
Set up and audit environment variable management — create .env.example, add startup validation, separate secrets from config, and document every variable