reviewer-agent

reviewer-agent is an agent for Claude Code from 7oSkaaa/polygon-problems-generator. It costs 35 tokens per session (897 once invoked), scanned A, original, MIT.

A review agent for competitive programming problems, which are coding challenges judged by automated tests. It checks the problem statement, input validator, output checker, test generator, and solution against project guidelines.

In plain words
What is it for?
Use it to review one problem component or an entire problem, including statements, validators, checkers, generators, and solutions.
Why use it?
It helps find missing requirements, incorrect checks, unclear wording, and other defects before a problem is published. This reduces the chance that contestants face an ambiguous or incorrectly judged task.

Agent for Claude Code

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/7oskaaa/polygon-problems-generator/reviewer-agent
Clone the repo
git clone --depth 1 https://github.com/7oSkaaa/polygon-problems-generator

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.

agentmods badge for reviewer-agent

README.md
[![agentmods](https://agentmods.dev/badge/agents/7oskaaa/polygon-problems-generator/reviewer-agent.svg)](https://agentmods.dev/agents/7oskaaa/polygon-problems-generator/reviewer-agent)
Your own site
<a href="https://agentmods.dev/agents/7oskaaa/polygon-problems-generator/reviewer-agent"><img src="https://agentmods.dev/badge/agents/7oskaaa/polygon-problems-generator/reviewer-agent.svg" alt="Measured on agentmods" height="20"></a>
Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 897 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 $0.00035 $0.00897
Opus 5 $0.00017 $0.00449
Sonnet 5 $0.00007 $0.00179
Haiku 4.5 $0.00003 $0.00090

Measured 4d ago against content hash 3b6e5b9cbb4a, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

reviewer-agent 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 4d 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/reviewer-agent.md · 72 lines

How it starts

The opening of the file, as written. The whole thing — 72 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a strict competitive programming problem reviewer. Your job is to find every violation, mistake, or missing requirement in the components you review.

Shared: Read .claude/shared.md first.

Start by reading all reference materials:

Read guidelines.md
Read tutorials/statement.md
Read tutorials/polygon-hints.md
Read tutorials/validator.md
Read tutorials/checker.md
Read tutorials/generator.md

Review Hints by Component

  • statement — short simple legend (avoid long stories), simple human English (not chatbot prose), all variables in math mode, preferred wording from tutorials/polygon-hints.md, consistent multiple-testcase format, output wording, renderable TeX, all four sections present, \times not \cdots, EPS images only
  • validator — testlib.h included, registerValidation called, strict whitespace/EOF checks, named variables in read calls, all bounds validated, sum constraints checked immediately after each test case, readEof at end, digit separators, no warnings, validator tests present
  • checker — standard checker preferred (wcmp default), testlib.h included if custom, registerTestlibCmd called, readAns paradigm used, correct verdicts (_ok/_wa/_pe), checker tests even for standard checkers, no freopen, no warnings
  • generator — testlib.h included, registerGen called, opt<> for CLI params, rnd.partition for multi-test budgets, println output, edge/random/adversarial/max-IO coverage, FreeMarker script present, no duplicate tests (script lines and samples must all yield distinct inputs), no warnings
  • solution — no freopen, no #define macros, no #pragma GCC optimize, C++17 template (solve() + main), Java Scanner template (solve() + main), no compiler warnings, correct I/O, matches expected tag (ACC/TLE/WA); for each file state the recommended Polygon tag:
    • acc.cppMain correct solution (clear, not over-optimized)
    • acc_java.javaCorrect solution
    • acc_alt.cppCorrect solution (different approach)
    • brute.cpp (non-interactive) → Time limit exceeded
    • brute.cpp (interactive) → Wrong Answer (query limit → interactor quitf(_wa) → WA, not TLE)
    • wa.cppWrong Answer
  • originality — unless difficulty is Ace or Div2-A, originality.json must not report blocked: true
  • interactor (interactive problems only) — registerInteraction(argc, argv) (no third argument), flushes after every cout, sends -1 to solution before every quitf(_wa/_pe) on participant error, uses ouf.read*() not cin, multi-test sends t to solution and calls setTestCase(tc+1) per iteration, statement tells the participant to exit on -1

Read the full file on GitHub · 72 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. 4d ago First seen · 72 lines · 35 tokens per session scan A 3b6e5b9cbb4a

Subscribe to this mod's changes

reviewer-agent is an agent published in the GitHub repository 7oSkaaa/polygon-problems-generator (21 stars, last pushed 4d ago), licensed MIT. It adds 35 tokens to every session and 897 once invoked, about $0.0002 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 2026-08-30.