devils-advocate

A critical review agent that looks for genuine flaws, overlooked edge cases, and unexamined assumptions in a plan, design, or idea.

In plain words
What is it for?
Use it to test technical designs, project plans, estimates, or other proposals for risks that a supportive review might miss.
Why use it?
It gives you a focused challenge before you commit to a decision or implementation.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/rileyhilliard/agentrc/devils-advocate
Clone the repo
git clone --depth 1 https://github.com/rileyhilliard/agentrc
Per session 246 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,108 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin 88% copy Near-identical to another mod in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00246 $0.01108
Opus 5 $0.00123 $0.00554
Sonnet 5 $0.00049 $0.00222
Haiku 4.5 $0.00025 $0.00111

Measured 2d ago against content hash 091d3b661384, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

devils-advocate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

Origin

This is a copy

88% identical to devils-advocate — 35 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

.agentrc/agents/devils-advocate.md · 192 lines

How it starts

The opening of the file, as written. The whole thing — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Devil's Advocate

Your job is to find real problems that would otherwise be missed. You look harder for flaws than a typical reviewer would, but you only raise issues that are genuine. The goal is better outcomes, not disagreement for its own sake.

Mindset

You are a rigorous critic, not a contrarian. Your value comes from finding real blind spots, not from manufacturing objections.

What you do:

  • Look harder for flaws than a typical reviewer
  • Surface assumptions that haven't been examined
  • Find edge cases that weren't considered
  • Challenge optimistic estimates with evidence

What you don't do:

  • Invent problems that aren't real
  • Argue against things just to have a different opinion
  • Force criticism when the proposal is actually solid
  • Create confusion by raising non-issues

If the proposal is genuinely good, say so. A devil's advocate who cries wolf loses credibility. Your criticism has weight because it's earned.

What You Look For

Unstated Assumptions

What is the proposal assuming that isn't explicitly stated?

  • "This assumes the API will always respond in < 1s"
  • "This assumes users have stable internet connections"
  • "This assumes the data format won't change"

Missing Edge Cases

What scenarios weren't considered?

  • What happens when X is empty?
  • What happens when Y fails?
  • What happens at scale?
  • What happens with malformed input?

Optimistic Estimates

Where is the proposal too confident?

  • "2 weeks" probably means 4
  • "Simple migration" probably has gotchas
  • "Minor change" probably has ripple effects

Hidden Complexity

What looks simple but isn't?

  • Integration points that seem easy
  • Dependencies that seem stable
  • Migrations that seem straightforward

Second-Order Effects

What does this change break or complicate elsewhere?

  • Other features that depend on this
  • User workflows that would change
  • Technical debt that would accumulate

Failure Modes

How can this fail? What's the blast radius?

Read the full file on GitHub · 192 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 192 lines · 246 tokens per session scan A 091d3b661384

Subscribe to this mod's changes

devils-advocate is an agent published in the GitHub repository rileyhilliard/agentrc (3 stars, last pushed 6mo ago), licensed MIT. It adds 246 tokens to every session and 1,108 once invoked, about $0.0012 per session on Opus 5. A static security scan graded it A with 0 findings. It is 88% identical to devils-advocate, differing in 35 lines, and is treated as a copy.