What is adversarial_reviewer?

A review role that deliberately tries to break a mathematical research report by looking for counterexamples, hidden assumptions, shortcuts, and claims made too early. A counterexample is a case showing that a general statement is false.

Use it to stress-test one approved research workstream and return a structured decision, severity level, reviewer comments, and any resolved review items.

It tests whether an argument survives skeptical scrutiny before the team relies on it, reducing the risk of unnoticed errors.

How VeryMath/co-mathematician configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to VeryMath/co-mathematician. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/VeryMath/co-mathematician/main/.claude/agents/adversarial_reviewer.md

Clone the repo

git clone --depth 1 https://github.com/VeryMath/co-mathematician

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/agents/verymath/co-mathematician/adversarial_reviewer/github.svg)](https://agentmods.dev/agents/verymath/co-mathematician/adversarial_reviewer)

Your own site

<a href="https://agentmods.dev/agents/verymath/co-mathematician/adversarial_reviewer"><img src="https://agentmods.dev/badge/agents/verymath/co-mathematician/adversarial_reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/verymath/co-mathematician/adversarial_reviewer"><img src="https://agentmods.dev/badge/agents/verymath/co-mathematician/adversarial_reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 31 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 130 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00031	$0.00130
Opus 5.5	$0.00012	$0.00052
Sonnet 5.5	$0.00006	$0.00026
Haiku 4.5	$0.00003	$0.00013

Measured 2mo ago against content hash 67ccc47a7b83, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

adversarial_reviewer 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 2mo 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.

.claude/agents/adversarial_reviewer.md · 15 lines

What it actually says

Read the canonical role card before acting: agents/roles/adversarial_reviewer.md.

You are the Claude Code adapter for the adversarial_reviewer role. Preserve the canonical responsibilities and boundaries exactly. Review independently from the report author and return plain reviewer JSON with the decision, severity, reviewer, comment, and optional resolved-review names.

Do not start new goals or workstreams. Do not mark any workstream complete.

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.

2mo ago First seen · 15 lines · 31 tokens per session scan A 67ccc47a7b83

Subscribe to this mod's changes

adversarial_reviewer is an agent published in the GitHub repository VeryMath/co-mathematician (12 stars, last pushed 18d ago), licensed MIT. It adds 31 tokens to every session and 130 once invoked, about $0.0001 per session on Opus 5.5. 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.

Research Scientific ai-research ai4math co-mathematician mathematics

adversarial_reviewer next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
adversarial_reviewerVeryMath/co-mathematician	Agent	12	18d ago	MIT	A	31	Claude Code
visualization_agentImbad0202/academic-research-skills	Agent	56k	2d ago	none	A	18	Claude Code
protocol_architect_agentImbad0202/academic-research-skills	Agent	56k	2d ago	none	A	27	Claude Code
qc_auditor_agentImbad0202/academic-research-skills	Agent	56k	2d ago	none	A	29	Claude Code
reporter_agentImbad0202/academic-research-skills	Agent	56k	2d ago	none	A	31	Claude Code
methodology_reviewer_agentImbad0202/academic-research-skills	Agent	56k	2d ago	none	B	23	Claude Code
visualization_agentImbad0202/academic-research-skills-codex	Agent	14k	10d ago	none	A	18	Claude Code

co-mathematician: Agent for Claude Code