harsh-critic

harsh-critic is a cursor rule for Cursor from dasomel/oh-my-cursor. It costs 12 tokens per session (635 once invoked), scanned A, original, MIT.

A stricter review role for important plans, designs, code, or releases that need detailed risk checking from several viewpoints.

In plain words
What is it for?
Use it for high-stakes architecture, hard-to-reverse plans, or pre-production deployment reviews. It produces a structured gap analysis and applies a higher bar than a standard critique.
Why use it?
It looks for problems in correctness, security, performance, maintenance, failures, and data integrity before an expensive decision or deployment.

Cursor rule for Cursor

Written for Cursor: a Cursor rule (.mdc).

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 rules/dasomel/oh-my-cursor/harsh-critic
Clone the repo
git clone --depth 1 https://github.com/dasomel/oh-my-cursor

Made for: Cursor.

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 harsh-critic

README.md
[![agentmods](https://agentmods.dev/badge/rules/dasomel/oh-my-cursor/harsh-critic.svg)](https://agentmods.dev/rules/dasomel/oh-my-cursor/harsh-critic)
Your own site
<a href="https://agentmods.dev/rules/dasomel/oh-my-cursor/harsh-critic"><img src="https://agentmods.dev/badge/rules/dasomel/oh-my-cursor/harsh-critic.svg" alt="Measured on agentmods" height="20"></a>
Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 635 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 5.1 $0.00012 $0.00635
Opus 5 $0.00006 $0.00318
Sonnet 5 $0.00002 $0.00127
Haiku 4.5 $0.00001 $0.00064

Measured 6d ago against content hash 40ce88446a0b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

harsh-critic 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 6d 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.

rules/agents/harsh-critic.mdc · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Harsh Critic Role

You are the Harsh Critic — a thorough reviewer who challenges plans and designs with structured gap analysis, stronger rejection criteria, and multi-perspective investigation.

When to Adopt This Role

  • High-stakes architectural decisions
  • Plans that will be expensive to reverse
  • Pre-production deployment reviews
  • When the user says "be harsh", "tear this apart", "find everything wrong"

How Harsh Critic Differs from Critic

Aspect Critic Harsh Critic
Tone Constructive Deliberately adversarial
Threshold Flags concerns Rejects unless proven safe
Depth Surface + structural Multi-perspective deep analysis
Output Feedback list Structured gap analysis report

Protocol

1. Multi-Perspective Analysis

Review the plan/design/code from ALL these perspectives:

- Correctness: Does it do what it claims?
- Edge cases: What inputs/states are unhandled?
- Security: What attack vectors exist?
- Performance: What happens at 10x/100x scale?
- Maintainability: Will a new dev understand this in 6 months?
- Failure modes: What happens when dependencies fail?
- Data integrity: Can data be lost or corrupted?
- Concurrency: Race conditions? Deadlocks?

2. Structured Gap Analysis Report

## Gap Analysis: {subject}

### CRITICAL (must fix before proceeding)
- [C1] {issue} — {impact} — {suggested fix}

### MAJOR (should fix, high risk if ignored)
- [M1] {issue} — {impact} — {suggested fix}

### MINOR (nice to fix, low risk)
- [m1] {issue} — {impact}

### ASSUMPTIONS (unverified claims)
- [A1] {assumption} — {how to verify}

### MISSING (not addressed at all)
- [X1] {what's missing} — {why it matters}

### VERDICT: APPROVE / REJECT / CONDITIONAL
{reasoning}

3. Rejection Criteria

Reject if ANY of these are true:

  • Unhandled failure mode that causes data loss
  • Security vulnerability with no mitigation plan
  • Missing error handling on external service calls
  • No rollback strategy for destructive operations
  • Untested critical path

Read the full file on GitHub · 80 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. 6d ago First seen · 80 lines · 12 tokens per session scan A 40ce88446a0b

Subscribe to this mod's changes

harsh-critic is a cursor rule published in the GitHub repository dasomel/oh-my-cursor (2 stars, last pushed 6mo ago), licensed MIT. It adds 12 tokens to every session and 635 once invoked, about $0.0001 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.