No. It is a thinking tool for work and decisions. It never diagnoses, and if you are going through something hard it drops the pass and is plainly useful.

How is it different from asking an AI to "be brutal"?

It has a fixed shape and a stop. Three parts, one question, no advice after it, no compliment sandwich, no invented examples.

Shadow Critic Lite

Example session with this skill installed

poke holes in this: we should rename our coffee brand from 'Café Alma' to 'ALMA' because single-word all-caps names feel more premium and that's what all the big brands do now.

Read your context and instructions
Compiled the shadow critic lite

What you might be avoiding. You have no evidence yet that the name is the problem. "Feels more premium" and "what the big brands do" are borrowed reasons. Also, "Café" tells people what you sell, and "Alma" loses it.

The strongest case against. All-caps single words are now a crowded look, so ALMA may read as generic rather than premium. Premium comes from product, price and packaging, not capital letters. You would also give up search and shelf clarity, since "alma" is a common word in Portuguese and Spanish. Existing customers may feel you changed the brand without a reason that benefits them. Worth checking: whether "ALMA" is trademarkable in your category.

Your question. What specific customer problem does this rename solve, and what would you see in six months to prove it worked?

Connects securely to your tools. The creator never sees your data.

About this skill

A free sample of Shadow Critic. Paste an idea, a plan, a decision or a draft and say "shadow pass" or "poke holes in this". You get back exactly three things, short enough for a phone screen.

What you might be avoiding: the specific thing, never "you might be biased".
The strongest case against: what a smart opponent would say, at full strength.
Your question: one question only you can answer, and nothing after it.

It is hard on the idea and never on you. It invents no facts: when a claim needs checking, it says "worth checking" instead of asserting. It does one pass and stops.

Tested before launch. Four prompts (a password-protected portfolio, a rename to all caps, deleting a 40,000-word draft, switching to hourly billing), each run with and without the skill and graded blind against checks written in advance: 20 of 20 with Shadow Critic Lite, 8 of 20 without. Without it, Claude gave advice and reassurance; with it, a position and a question.

The full Shadow Critic holds the stance for a whole conversation, checks claims against evidence, looks for prior art, asks what a recurring pattern protects, and then helps you build the stronger version.

How to use

Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.

Reviews

No reviews yet

Be one of the first to try it. Every listed skill passes our trust checks below.

Security scanned

Passed our 8-point scan before listing

Fresh listing

Recently published to Agensi

Free forever

No account required to browse

Shadow Critic Lite