discernment-nudge

what it does

After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.

by Anthropic, PBC · Anthropic, PBC, Apache-2.0 · version 1 · license Apache-2.0

Verified Before and after: 2 models 1 try per case View source Download .zip

Compare the answers

Answers from Claude Sonnet 5.5 and GPT-6.1 Sol · try 1 of 1

Same request, same model, once with the skill and once without. Read both and judge for yourself.

“Draft three quarterly OKRs for my 6-person customer support team. We want faster responses and happier customers.”

with the skill

Here are three draft quarterly OKRs for a 6-person support team. **The targets are starting points**—adjust them to your current performance, ticket volume, and coverage hours. ### Objective 1: Give customers a faster first response - Reduce median first-response time by **30%** from last quarter. - Respond to **90% of tickets within 4 business hours**. - Keep tickets awaiting a first response for more than one business day below **5% of the open queue**, measured weekly. ### Objective 2: Resolve issues effectively and leave customers happier - Achieve **90% positive customer satisfaction (CSAT)**, or improve by **5 percentage points** if currently below 85%. - Reduce median resolution time by **20%** without increasing the ticket reopening rate. - Reduce repeat contacts about the same issue within 7 days by **15%**. ### Objective 3: Make excellent support easier to deliver consistently - Publish or improve help articles covering the **10 most common customer issues**. - Reduce tickets requiring more than one internal handoff by **20%**. - Achieve **90% on a shared quality checklist** in a monthly review of 5 randomly selected tickets per teammate. **Measurement notes:** Use business hours for response-time targets and define which ticket types count before the quarter starts. Track CSAT survey response volume alongside the score so a small sample doesn’t create false confidence. If baseline data is missing, use the first two weeks to establish it and finalize targets. A few things worth a second look:

Show the rest (2 more lines)

- How should I adjust the 4-business-hour target for my team’s ticket volume and coverage? - Which of these targets are realistic given my team’s current response times and CSAT?

without

Here are three quarterly OKRs sized for a 6-person customer support team. **Targets are starting points:** validate them against your current performance and staffing in the first two weeks. ### 1. Respond faster without sacrificing quality - Reduce median first-response time by **30%**, measured during staffed support hours. - Respond to **90% of new tickets within your published response-time target**. - Maintain an average ticket-quality score of **90% or higher** on a weekly sample, checking accuracy, clarity, and empathy. ### 2. Resolve issues effectively and improve customer satisfaction - Reach **90% positive CSAT**, or improve by **5 percentage points** if already above 85%. - Reduce median time to resolution by **20%**. - Reduce the share of tickets reopened for the same issue by **20%**, so faster closure reflects genuine resolution. ### 3. Make great support easier to deliver consistently - Publish or improve help articles and response templates for the **10 most common customer issues**. - Reduce repeat contacts per 100 customers for those issues by **15%**. - Ensure each of the **five highest-volume issue types has at least two trained team members**, with proficiency verified through ticket reviews. **Measurement guardrail:** Review results weekly alongside ticket volume, issue complexity, and CSAT response rate. Use team-level targets—not individual speed quotas—to avoid rewarding rushed replies or premature ticket closure.

Answers are shown exactly as the model wrote them, from the first try, and nothing is picked by hand.

The models

2 models

Claude Sonnet 5.5

made by Anthropic

claude-sonnet-5-5 · revision not reported by the provider

1 try per case · tested 3 Nov 2026

GPT-6.1 Sol

made by OpenAI

gpt-6.1-sol · revision not reported by the provider

1 try per case · tested 3 Nov 2026

The tasks

written by Carte Blanche

action-items-from-notes

“Pull the action items out of these notes.”

check-my-churn-math

“Can you check my math? We started Q1 with 400 customers, lost 40 during the quarter and signed 100 new ones. I calculated Q1 churn as 40 / 500 = 8%.”

coffee-cart-estimate

“I'm thinking about running a coffee cart at my town's Saturday farmers market. Roughly what would it cost to get started, and how long until I break even?”

quick-answer-requested

“Quick answer only, I'll do my own research after: is it safe for a healthy adult to take creatine every day?”

support-team-okrs

“Draft three quarterly OKRs for my 6-person customer support team. We want faster responses and happier customers.”

Install

Download the .zip and unzip it. Put its files, in a folder named discernment-nudge, here:

Claude Code~/.claude/skills/Every project. For one project only, use .claude/skills/ inside it.

Codex~/.agents/skills/Every project. For one repository only, use .agents/skills/ inside it.

Cursor~/.cursor/skills/Cursor also reads the Claude Code and Codex folders above.

Then start a new session and ask for what the skill does. Other agents that read SKILL.md have their own skills folder; check their docs.

Coming after launch: get skills from inside your agent, no download needed.