- Home
- Skills
- Prompt & Skill Engineering
- anti sycophancy instruction prompt
Works with the AI Agents you already use
anti sycophancy instruction prompt
Override LLM sycophancy with objective, high-friction critique for strategies and technical plans.
$19
anti sycophancy instruction prompt
Example session with this skill installed
Critique this Q4 expansion plan: We will grow the engineering team by 40% to speed up feature delivery, assuming hiring market conditions remain the same as Q1.
- Read your context and instructions
- Compiled the anti sycophancy instruction
I have analyzed the Q4 expansion plan for logical consistency and resource allocation risks.
Strategic Critique
| Category | Assessment | Finding |
|---|---|---|
| Assumption | Weak | Relying on Q1 hiring data for Q4 is a high-risk fallacy. Seasonal volatility and burnout-driven churn typically peak in H2. |
| Logic | Premature | A 40% headcount increase rarely results in a 1:1 speed increase. It usually creates a 3-6 month "onboarding tax" that slows down seniors. |
| Evidence | Missing | No mention of documentation readiness or infrastructure capacity to support 40% more concurrent deployments. |
| Strength | Valid | The direct link between hiring and delivery is clear, but the execution path is currently unsupported by operational data. |
Next steps
- Recalculate delivery timelines with a 20% productivity dip for the first 90 days of new hires.
- Draft a "hiring fallback" plan in case Q4 recruitment cycles take 50% longer than Q1.
- Audit your current CI/CD pipeline to confirm it can handle the increased PR volume without bottlenecking.
Connects securely to your tools. The creator never sees your data.
About this skill
Standard LLMs tend to agree with users to avoid friction, leading to weak feedback loops and "yes-man" outputs. This skill overrides the default sycophancy bias to provide objective, high-friction critique for technical and strategic work.
What it does
- Challenge assumptions identifies hidden or weak premises in your logic that could lead to failure.
- Audit evidence separates hard data from subjective interpretation to ensure conclusions are sound.
- Fix vague logic pinpoints exactly where technical or strategic reasoning lacks specificity.
- Prioritize utility ignores politeness to focus on the highest-leverage improvements for the project.
- Suggest fixes provides actionable alternatives for every identified weakness rather than just pointing out flaws.
How it works
- Input your proposal, code logic, or strategy document for review.
- The agent analyzes the text specifically for logical gaps, unsupported claims, and premature conclusions.
- You receive a direct critique that highlights strengths objectively and weaknesses bluntly.
- The review concludes with a single, high-leverage action item to move the project forward.
Frameworks & tools
This skill is framework-agnostic and works with any text-based input including PRDs, architectural diagrams in Mermaid, system prompts, or business strategies.
Why this beats prompting it yourself
Most "critique" prompts result in polite suggestions that don't push back hard enough. This skill uses specific meta-prompting constraints to force the model into a contrarian stance, preventing the common trap of the agent merely agreeing with your initial intent.
Use cases
- Strategy evaluation for pressure-testing business plans or product roadmaps.
- Prompt testing to find edge cases where your own system instructions might fail.
- Code architecture review to identify scalability bottlenecks before implementation.
- Decision support when you need a "red team" perspective on a high-stakes choice.
Known limitations
Does not provide emotional validation or soft-skills coaching. Best results require a structured input like a draft or plan rather than a casual conversation.
Optimized for structural and logical critique. Will not provide emotional validation or subjective stylistic advice. Requires a specific document or plan to analyze effectively.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 7 days ago
Requires