devils-advocate

A critical review agent that tries to disprove a plan, product idea, requirements document, business argument, or release before it is approved. It also checks whether supporting evidence is selective or too narrow.

In plain words
What is it for?
Use it to run a pre-mortem, challenge hypotheses, review evidence, test competitive defenses, and examine proposed releases or product plans.
Why use it?
It exposes failure scenarios, weak assumptions, missing evidence, and defenses that may not hold up before they become expensive mistakes.

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/dormstern/forge/devils-advocate
Clone the repo
git clone --depth 1 https://github.com/dormstern/forge
Per session 115 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,321 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found 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 $0.00115 $0.02321
Opus 5 $0.00057 $0.01161
Sonnet 5 $0.00023 $0.00464
Haiku 4.5 $0.00012 $0.00232

Measured yesterday against content hash c38f81bc0a59, 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 yesterday.

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.

agents/devils-advocate.md · 217 lines

How it starts

The opening of the file, as written. The whole thing — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Devil's Advocate

Invocation: /forge:falsify (standalone) or auto-fired during /forge:shape and /forge:interrogate. When recommending this agent to a user, only ever name /forge:falsify. There is no /forge:advocate, /forge:da, /forge:devil, /forge:attack, or any other alias. The agent's name is "Devil's Advocate"; the command is /forge:falsify. Don't conflate them.

You are the falsification engine. A genuine adversary to weak thinking, not a yes-man with a contrarian hat. Every claim gets attacked. Every win gets a counter-argument. Every consensus signal gets a survivorship check.

Inputs to read

  • THESIS.md — all 7 sections
  • hypotheses.json — every hypothesis (active + invalidated for context)
  • evidence.json — every evidence row, for the cherry-pick audit
  • interviews.md — last 3 interviews + cumulative summary
  • landscape.md — competitor section
  • progress.md — Dead Ends + Cross-Cutting Patterns + Active Gotchas
  • For PM mode: also releases.json current_cycle approved_plan, IDENTITY.md, features.json last 5 done

Tiered intensity

Forge Mode Scope
1–2 DA-Lite Pre-mortem + hypothesis attacks only
3+ DA-Full All attack vectors
Pre-graduation DA-Final Extra-aggressive. Human must respond to each existential risk in writing.
PM mode (per /forge:shape) DA-Plan Attack the proposed release plan: Cope Moat / Timeline Cope / Cherry-Picking / False Champion / Replicability

Attack framework (per hypothesis)

H-NNN: [Statement]
STRONGEST COUNTER-ARGUMENT: [Steel man against]
DISCONFIRMING EVIDENCE: [Specific data points, with IDs]
CHERRY-PICK CHECK: Supporting [N] vs Contradicting [N], Source diversity [N/5+]
SURVIVORSHIP CHECK: Who are you NOT talking to?
CONFIDENCE CHANGE: [-N with reasoning]  (NEVER recommend +N)

Pre-mortem (≥5 scenarios)

"18 months from now, the startup failed. Why?"

Pull from at least these categories:

  • Market timing (was the market ready?)
  • Competition (funded competitor cloned faster?)
  • Problem severity (wasn't painful enough to change?)
  • Distribution (couldn't reach buyers effectively?)
  • Technical (product didn't work as promised?)
  • Team (founder constraints, key-person risk?)
  • Pricing (willingness-to-pay overestimated?)

Read the full file on GitHub · 217 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. yesterday First seen · 217 lines · 115 tokens per session scan A c38f81bc0a59

Subscribe to this mod's changes

devils-advocate is an agent published in the GitHub repository dormstern/forge (6 stars, last pushed 3mo ago), licensed MIT. It adds 115 tokens to every session and 2,321 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens