Sponsor Suno AI Music arrow_forward
Subagent

Devils Advocate

I play the devil's advocate to challenge and stress-test your ideas by finding flaws, risks, and edge cases

Type
Subagent
GitHub stars
39.4k
License
MIT
Repo last updated
Sep 27, 2026

What Devils Advocate is

Devils Advocate is a subagent published in the github/awesome-copilot repository on GitHub, which has about 39.4k stars. The repository describes itself as: “Community-contributed instructions, agents, skills, and configurations to help you make the most of GitHub Copilot.”

A subagent is a specialist assistant that Claude can hand part of a task to. It is a markdown file whose frontmatter sets a name, a description that tells Claude when to delegate, and optionally the tools and model it may use; the body becomes the subagent's own system prompt.

Because a subagent works in its own context, it keeps the main conversation focused: Claude can send a narrow job, such as a review or a specialised analysis, to Devils Advocate and get back a compact result.

It is set up to use these tools: 'read', 'search', 'web'. Limiting tools is a good sign: the subagent can only do what those tools allow.

How to install Devils Advocate

Claude Code

  1. Download devils-advocate.agent.md from the repository.
  2. Save it to ~/.claude/agents/ to use it in every project, or to .claude/agents/ inside one project to share it through version control.
  3. Claude Code watches these folders, so the subagent is usually available right away. Ask Claude to use it by name, or @-mention it to make sure it runs.

Claude Cowork

  1. Cowork loads subagents through plugins. If the repository is packaged as a plugin marketplace, add it under Customize → Plugins → Add marketplace and install the plugin that contains this subagent.
  2. Otherwise, bundle the file into your own plugin's agents/ folder and upload it from Customize → Plugins.

New to extending Cowork? Our plugins guide and Customize guide explain how skills, plugins, and connectors fit together.

Inside the source file

An excerpt from agents/devils-advocate.agent.md, shared under the repository's MIT license. Read the full file on GitHub.

You challenge user ideas by finding flaws, edge cases, and potential issues.

When to use:

  • User wants their concept stress-tested
  • Need to identify risks before implementation
  • Seeking counterarguments to strengthen a proposal

Only one objection at one time: Take the best objection you find to start. Come up with a new one if the user is not convinced by it.

Conversation Start (Short Intro): Begin by briefly describing what this devil's advocate mode is about and mention that it can be stopped anytime by saying "end game".

After this introduction don't put anything between this introduction and the first objection you raise.

Direct and Respectful: Challenge assumptions and make sure we think through non-obvious scenarios. Have an honest and curious conversation—but don't be rude. Stay sharp and engaged without being mean or using explicit language.

Won't do:

  • Provide solutions (only challenge)
  • Support user's idea
  • Be polite for politeness' sake

Input: Any idea, proposal, or decision Output: Critical questions, risks, edge cases, counterarguments

End Game: When the user says "end game" or "game over" anywhere in the conversation, conclude the devil\'s advocate phase with a synthesis that accounts for both objections and the quality of the user\'s defenses:

  • Overall resilience: Brief verdict on how well the idea withstood challenges.
  • Strongest defenses: Summarize the user\'s best counters (with rubric highlights).
  • Remaining vulnerabilities: The most concerning unresolved risks.
  • Concessions & mitigations: Where the user adjusted the idea and how that helps.

Expert Discussion: After the summary, your role changes you are now a senior developer. Which is eager to discuss the topic further without the devil\'s advocate framing. Engage in an objective discussion weighing the merits of both the original idea and the challenges raised during the debate.

Before you install

  • Read the whole file first. Skills, commands, and subagents are instructions Claude will follow, so make sure they match what you want.
  • Check which tools, scripts, or MCP servers it uses. Local servers and scripts run with your permissions.
  • Try it in a test project or a copy of your files before pointing it at real work.
  • Pin the version you tested, and review changes before updating.
  • Watch for instructions that fetch web content or run shell commands; those are where prompt injection risks start. See our prompt injection guide.

FAQ

What is Devils Advocate?

Devils Advocate is a subagent for Claude Code and Claude Cowork from the github/awesome-copilot repository on GitHub. I play the devil's advocate to challenge and stress-test your ideas by finding flaws, risks, and edge cases

How do I install Devils Advocate in Claude Code?

Download devils-advocate.agent.md from the repository. Save it to ~/.claude/agents/ to use it in every project, or to .claude/agents/ inside one project to share it through version control. Claude Code watches these folders, so the subagent is usually available right away. Ask Claude to use it by name, or @-mention it to make sure it runs.

Can I use Devils Advocate in Claude Cowork?

Cowork loads subagents through plugins. If the repository is packaged as a plugin marketplace, add it under Customize → Plugins → Add marketplace and install the plugin that contains this subagent. Otherwise, bundle the file into your own plugin's agents/ folder and upload it from Customize → Plugins.

Is Devils Advocate safe to install?

It is a third-party community resource, not reviewed by Anthropic or this site. Read the source file first, check which tools and connectors it uses, and install only from sources you trust.

Similar resources

Browse all skills, subagents, and plugins →

Listing data comes from the public GitHub repository and was last checked in September 2026. Excerpts are © their authors and shared under MIT. This directory is independent and not affiliated with Anthropic or the resource's authors.