$ loading_
Create, fix, and optimize VS Code agent customization files and workflows.
The material indicates a prompt/workflow-style skill from an open-source repository, with no required secrets and no declared remote endpoints; its functionality is mainly about guiding or editing VS Code customization files, so overall risk is low. Some caution is still warranted because the description mentions file-system tools and subagents for local file operations, which implies ordinary local data access and execution capabilities.
The material explicitly states that no keys or environment variables are required, and there is no request for API tokens, account credentials, or other sensitive authentication data, so credential exposure and abuse risk appears low.
Both the system checks and the material indicate no remote endpoints, and there is no stated transfer of user data to external services; the docs mention official reference links, but do not show inherent outbound data exfiltration by the skill itself.
The description mentions using file-system tools, an ask-questions tool, and subagents for codebase exploration; this implies some local tool invocation and agent orchestration capability, but there is no evidence of unrelated system command execution or unusually elevated privileges.
Its core purpose is to create and modify workspace-level or user-level VS Code customization files, which clearly involves local file read/write access; the stated access scope is broadly consistent with the declared functionality, with no sign of excessive bulk data collection.
The source points to Microsoft's open-source vscode repository on GitHub, making the code auditable and materially lowering supply-chain risk. Although community adoption and maintenance metadata are incomplete and the license is not stated here, there is no evidence of closed-source delivery, an unknown publisher, or clear injection red flags.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "agent-customization" skill from askskill: 1. Download https://raw.githubusercontent.com/microsoft/vscode/main/extensions/copilot/assets/prompts/skills/agent-customization/SKILL.md 2. Save it as ~/.claude/skills/agent-customization/SKILL.md 3. Reload skills and tell me it's ready
Create a copilot-instructions.md for my VS Code project. Include: TypeScript first, tests must run before commit, API code goes under src/api, responses should be concise, and explain which files these rules apply to.
A well-structured customization file with rules, scope, and save-ready formatting.
Help me figure out why my .instructions.md is not being applied. Analyze YAML frontmatter, applyTo patterns, file naming, directory placement, and rule conflicts, then suggest fixes.
A diagnosis listing likely causes, priority fixes, and corrected example configuration.
Design a custom agent mode for frontend refactoring. It should first read the component directory, identify duplicated logic, restrict edits to UI-layer files only, and structure output into plan, change suggestions, and risk notes.
A practical agent customization plan with mode definition, tool restrictions, execution steps, and example file content.
| Primitive | When to Use |
|---|---|
| agent instructions | Always-on, applies everywhere in the project |
| File Instructions | Explicit via applyTo patterns, or on-demand via description |
| MCP | Integrates external systems, APIs, or data |
| Hooks | Deterministic shell commands at agent lifecycle points (block tools, auto-format, inject context) |
| Custom Agents | Subagents for context isolation, or multi-stage workflows with tool restrictions |
| Prompts | Single focused task with parameterized inputs |
| Skills | On-demand workflow with bundled assets (scripts/templates) |
Consult the reference docs for templates, domain examples, advanced frontmatter options, asset organization, anti-patterns, and creation checklists. If the references are not enough, load the official documentation links for each primitive.
| Type | File | Location | Reference |
|---|---|---|---|
| agent instructions | copilot-instructions.md, AGENTS.md | .github/ or root | Link |
| File Instructions | *.instructions.md | .github/instructions/ | Link |
| Prompts | *.prompt.md | .github/prompts/ | Link |
| Hooks | *.json | .github/hooks/ | Link |
| Custom Agents | *.agent.md | .github/agents/ | Link |
| Skills | SKILL.md | .github/skills/<name>/, .agents/skills/<name>/, .claude/skills/<name>/ | Link |
User-level: {{VSCODE_USER_PROMPTS_FOLDER}}/ (*.prompt.md, *.instructions.md, *.agent.md; not skills)
Customizations roam with user's settings sync
If you need to explore or validate patterns in the codebase, use a read-only subagent. If the ask-questions tool is available, use it to interview the user and clarify requirements.
Follow these steps when creating any customization file.
Ask the user where they want the customization:
.github/ folder{{VSCODE_USER_PROMPTS_FOLDER}}/Use the Decision Flow above to select the appropriate file type based on the user's need.
Create the file directly at the appropriate path:
After creating:
--- markers)description is present and meaningfulInstructions vs Skill? Does this apply to most work, or specific tasks? Most → Instructions. Specific → Skill.
Skill vs Prompt? Both appear as slash commands in chat (type /). Multi-step workflow with bundled assets → Skill. Single focused task with inputs → Prompt.
Skill vs Custom Agent? Same capabilities for all steps → Skill. Need context isolation (subagent returns single output) or different tool restrictions per stage → Custom Agent.
Hooks vs Instructions? Instructions guide agent behavior (non-deterministic). Hooks enforce behavior via shell commands at lifecycle events like PreToolUse or PostToolUse — they can block operations, require approval, or run formatters deterministically. Hooks can be defined in standalone .json files (see hooks reference) or inline in custom agent frontmatter via the hooks attribute (see agents reference).
…
Validate Azure DevOps pipeline changes and troubleshoot builds and YAML faster.
Update the GitHub Copilot CLI or SDK to a newer version.
Upgrade Anthropic SDKs, migrate versions, and fix dependency or typing issues.
Generate or update chat customization files for AI coding agents.
Find and read Code OSS dev build logs for faster debugging.
Merge session branch changes back into the base branch cleanly.
Integrate skills, MCP servers, and agent configs to strengthen coding agents.
Mine recurring coding-agent workflows and refine skills for reuse and publication.
Extend AI coding agents with curated skills for development and professional workflows.
Configure and manage agents, skills, prompts, and integrations in the editor.
Create or update skills that extend Codex with knowledge, workflows, and tools.
Create, refine, validate, and restructure AgentSkills and SKILL.md files.