Skip to content
Back to skills

1781 Copilot Instructions 6f5dd2aa

ASecurity

> VS Code Copilot-specific orchestration instructions. > For general project conventions, build commands, and code style, see the root `AGENTS.md`.

  • 9 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added October 11, 2026
devopspythongobashazureterraformgitapisecuritydocumentation

Works with

  • vscode
  • cli
  • api
  • mcp

Security analysis

A100/100

Scanned October 11, 2026

npx -y skills add tools-only/X-Skills --skill 1781-copilot-instructions_6f5dd2aa --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of 1781 Copilot Instructions 6f5dd2aa?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for 1781 Copilot Instructions 6f5dd2aa
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tools-only-1781-copilot-instructions-6f5dd2aa/badge)](https://www.skillsdirectory.com/skills/tools-only-1781-copilot-instructions-6f5dd2aa)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

SKILL.md
# Agentic InfraOps - Copilot Instructions

> VS Code Copilot-specific orchestration instructions.
> For general project conventions, build commands, and code style, see the root `AGENTS.md`.

## Quick Start

1. Enable subagents: `"github.copilot.chat": { "customAgentInSubagent": { "enabled": true } }`
2. Open Chat (`Ctrl+Shift+I`) → Select **InfraOps Conductor** → Describe your project
3. The Conductor guides you through all 7 steps with approval gates

## 7-Step Workflow

| Step | Agent                                                                      | Output                                                                                                    | Review | Gate       |
| ---- | -------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- | ------ | ---------- |
| 1    | Requirements                                                               | `01-requirements.md`                                                                                      | 1x     | Approval   |
| 2    | Architect                                                                  | `02-architecture-assessment.md` + cost estimate                                                           | 3x+1x  | Approval   |
| 3    | Design (opt)                                                               | `03-des-*.{py,png,md}`                                                                                    | —      | —          |
| 4    | IaC Plan (Bicep: `05b-Bicep Planner` / Terraform: `05t-Terraform Planner`) | `04-implementation-plan.md` + governance + `04-dependency-diagram.py/.png` + `04-runtime-diagram.py/.png` | 1x+3x  | Approval   |
| 5    | IaC Code (Bicep: `06b-Bicep CodeGen` / Terraform: `06t-Terraform CodeGen`) | `infra/bicep/{project}/` or `infra/terraform/{project}/`                                                  | 3x     | Validation |
| 6    | Deploy (Bicep: `07b-Bicep Deploy` / Terraform: `07t-Terraform Deploy`)     | `06-deployment-summary.md`                                                                                | 1x     | Approval   |
| 7    | As-Built                                                                   | `07-*.md` documentation suite                                                                             | —      | —          |

All outputs → `agent-output/{project}/`. Context flows via artifact files + handoffs.
Review column = adversarial passes by `challenger-review-subagent` (3x = rotating lenses; 1x = comprehensive).

## Skills (Auto-Invoked by Agents)

| Skill                      | Purpose                                                               |
| -------------------------- | --------------------------------------------------------------------- |
| `azure-defaults`           | Regions, tags, naming, AVM, security, governance, pricing             |
| `azure-artifacts`          | Template H2 structures, styling, generation rules                     |
| `azure-bicep-patterns`     | Reusable Bicep patterns (hub-spoke, PE, diagnostics)                  |
| `azure-troubleshooting`    | KQL templates, health checks, remediation playbooks                   |
| `azure-diagrams`           | Python architecture diagram generation                                |
| `azure-adr`                | Architecture Decision Records                                         |
| `github-operations`        | GitHub issues, PRs, CLI, Actions, releases                            |
| `git-commit`               | Commit message conventions                                            |
| `docs-writer`              | Documentation generation                                              |
| `make-skill-template`      | Scaffold new Agent Skills from templates                              |
| `microsoft-docs`           | Query official Microsoft/Azure docs (requires Learn MCP)              |
| `microsoft-code-reference` | Verify SDK methods and find working code samples (requires Learn MCP) |
| `microsoft-skill-creator`  | Create hybrid skills for Microsoft technologies (requires Learn MCP)  |
| `terraform-patterns`       | Terraform HCL patterns (hub-spoke, PE, diagnostics, AVM pitfalls)     |

Agents read skills via: **"Read `.github/skills/{name}/SKILL.md`"** in their body.

## Chat Triggers

- If a user message starts with `gh`, treat it as a GitHub operation.
  Examples: `gh pr create ...`, `gh workflow run ...`, `gh api ...`.
- Automatically follow the `github-operations` skill guidance (MCP-first, `gh` CLI fallback) from `.github/skills/github-operations/SKILL.md`.

### GitHub MCP Priority (Mandatory)

- For issues and pull requests, always prefer GitHub MCP tools over `gh` CLI.
- Only use `gh` for operations that have no equivalent MCP write tool in the current environment.
- In devcontainers, do not run `gh auth` commands unless the user explicitly asks for CLI authentication troubleshooting.

## Key Conventions

See the root `AGENTS.md` for full conventions. Summary of VS Code-specific overrides:

- **AVM-first**: Always prefer Azure Verified Modules over raw Bicep/Terraform
- **Governance**: Always check `04-governance-constraints.md` for subscription-level Azure Policy

Full details in `.github/skills/azure-defaults/SKILL.md`.

### Terraform Conventions

Full details in `.github/skills/terraform-patterns/SKILL.md` and root `AGENTS.md`.

## Key Files

| Path                              | Purpose                                                 |
| --------------------------------- | ------------------------------------------------------- |
| `AGENTS.md`                       | Cross-agent project conventions and commands            |
| `.github/agents/*.agent.md`       | Agent definitions                                       |
| `.github/skills/*/SKILL.md`       | Reusable skill knowledge                                |
| `.github/instructions/`           | File-type rules (Bicep, Markdown, etc.)                 |
| `agent-output/{project}/`         | Agent-generated artifacts                               |
| `infra/bicep/{project}/`          | Bicep templates                                         |
| `mcp/azure-pricing-mcp/`          | Azure Pricing MCP server                                |
| `.vscode/mcp.json`                | MCP server configuration                                |
| `.vscode/infraops.toolsets.jsonc` | Workspace tool groups for interactive chat (8 toolsets) |
| `infra/terraform/{project}/`      | Terraform templates by project                          |
| `docs/tf-support/`                | Terraform support planning docs and prompts             |

## Validation

See `AGENTS.md` for full build and validation commands. Quick reference:

```bash
npm run validate:all
npm run lint:md
```

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.