本文へ移動
cccskills
無料GitHub で公開

feedback-delivery

Deliver feedback using Kim Scott's Radical Candor framework - care personally while challenging directly. Use when a leader needs to give difficult feedback, says "I need to tell someone something hard", "how do I give feedback without hurting them", "I've been avoiding a conversation", "my direct report isn't performing", "someone on my team has a blind spot", "I keep sugarcoating things", or "how do I be honest without being a jerk". Also trigger when someone describes being too nice (ruinous empathy) or too harsh (obnoxious aggression), or when they describe avoiding a conversation they know they need to have.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md7.3 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Overview

Based on "Radical Candor" by Kim Scott. The core idea: caring about someone does not mean protecting them from hard truths. Ruinous empathy - being "nice" by withholding difficult feedback - actually harms the person more than the feedback would. Radical Candor is the intersection of caring personally (genuine investment in the person) and challenging directly (honest, specific, timely feedback). Both axes matter. High care without directness is ruinous empathy. High directness without care is obnoxious aggression.

The Radical Candor Grid

                    CHALLENGE DIRECTLY
                         High
                          |
      Obnoxious     |    Radical
      Aggression    |    Candor
                    |
  CARE   Low ------+------ High   CARE
  Low               |
                    |
      Manipulative  |    Ruinous
      Insincerity   |    Empathy
                    |
                         Low

Most managers who think they are "being nice" are in Ruinous Empathy. Most managers who think they are "being direct" and come across as harsh are in Obnoxious Aggression. Radical Candor requires both axes high simultaneously.

Workflow

Step 1: Identify the Type of Feedback

Before delivering, know what kind of feedback this is:

  • Immediate course correction - behavior that needs to stop or change now
  • Pattern feedback - a recurring behavior that is limiting the person
  • Developmental feedback - honest assessment of a growth area for someone who wants to improve
  • Positive reinforcement - specific praise to anchor behavior you want repeated

Each type has the same framework but different urgency and framing.

Step 2: Check Your Care Baseline

Ask yourself: "Do I actually care about this person's success and growth?" If the honest answer is no, or if you are primarily irritated, wait. Feedback delivered from frustration lands as obnoxious aggression regardless of your words.

Signs you're ready:

  • You can state what success looks like for this person
  • You want them to succeed, not just to stop causing you problems
  • You can say one true positive thing about their work before the hard part

Step 3: Apply the SBI Framework

Use Situation-Behavior-Impact to make feedback specific and inarguable.

Situation - the specific context, time, place, meeting Behavior - the observable action (not interpretation, not character) Impact - the concrete effect on you, the team, or the work

Template:

In [situation], when you [specific behavior], the impact was [concrete consequence].

Examples:

Ruinous Empathy version: "You're doing great overall, just maybe think about being more communicative with the team."

Radical Candor version: "In the sprint planning meeting on Tuesday, when you didn't share the blocker on the auth service until the end of the meeting, the team had already committed to a two-week timeline that doesn't account for it. We had to replan with engineering for 90 minutes after."

The second version is specific enough that the person can change their behavior. The first version is noise.

Step 4: Deliver It

Timing: as close to the behavior as possible. Delayed feedback loses its anchor to the specific event and feels like a list of grievances.

Setting: 1:1, private. Never in front of others unless the behavior is currently happening and needs immediate correction.

Structure:

  1. State the observation (SBI)
  2. Ask for their perspective: "What was going on from your side?"
  3. Listen without defending your framing
  4. Agree on a specific change: "Going forward, what would look different?"

Do not:

  • Lead with "I feel" when you mean "you did" - this conflates emotional processing with feedback
  • Sandwich the feedback between compliments - the compliments dilute the signal
  • End with "but you're doing great overall" - this is the person's brain filing the feedback as "not serious"

Step 5: Follow Up

Feedback without follow-up is a lecture. Within 1-2 weeks:

  • Observe whether behavior changed
  • If it did - acknowledge it specifically: "I noticed you flagged the dependency issue in standup this morning. That's exactly what I was talking about."
  • If it didn't - have the conversation again. Repeat feedback is not nagging - it is accountability.

Step 6: Feedback Prep Worksheet

Before a hard conversation, fill this out:

Feedback Prep - [date]

Person: [your direct report or colleague]
Type: [course correction / pattern / developmental / positive]

Situation: [specific time and place]
Behavior: [observable action, no interpretation]
Impact: [concrete consequence on team, work, or you]

Their likely reaction: [defensive / dismissive / upset / receptive]
What I'll do if they get defensive: [specific response plan]

What success looks like for this person: [1-2 sentences]
Change I'm asking for: [specific, observable behavior going forward]

Follow-up plan: [when and what I'll look for]

Anti-Patterns

1. Ruinous empathy disguised as kindness Bad: Giving vague positive feedback because you don't want to upset someone, then being frustrated when they don't improve. Good: Specific, direct feedback is a gift. Withholding it because it's uncomfortable is prioritizing your comfort over their development.

2. Character attacks instead of behavior observations Bad: "You're not a team player." Good: "In the last three sprint reviews, you didn't mention the work your teammates contributed. The impact is that your teammates feel invisible." Feedback on character is inarguable and produces defensiveness. Feedback on behavior is specific and actionable.

3. Feedback in public Bad: Correcting someone's approach in a group meeting. Good: Pull them aside after. Public correction triggers shame, not reflection. The goal is change, not humiliation.

4. Saving it for the performance review Bad: Bringing up a pattern from six months ago in an annual review. Good: Six-month-old feedback is archaeology. Deliver it within a week of the behavior. The review should contain no surprises.

5. Skipping the follow-up Bad: Giving good feedback once and assuming the job is done. Good: People revert under pressure. One conversation is a data point. A pattern of consistent feedback is the intervention.

Quality Checklist

  • Feedback uses SBI format (specific situation, specific behavior, concrete impact)
  • Feedback is on observable behavior, not character or intent
  • Delivered in private, in a timely fashion (within a week of the event)
  • Asked for their perspective before concluding
  • Agreed on a specific, observable change
  • Follow-up date or trigger identified
  • Positive reinforcement is also specific (not just "good job")

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Rate any product, feature, or experience on the 11-star scale (Brian Chesky's Airbnb thought experiment). Use when user says "rate this experience", "11-star", "star rating", "experience audit", "how good is this", "experience rating", "product audit", "quality assessment", or wants to evaluate product quality and identify improvement paths. Also trigger when user wants to benchmark a feature, assess where a product stands, or map out what "great" looks like - even if they don't explicitly say "11-star".

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Generate A/B copy variants grounded in Eugene Schwartz's awareness and market-sophistication stages and John Caples' tested-headline discipline. Every variant declares its angle, the awareness stage it targets, and a falsifiable hypothesis, and the skill recommends which pair to test first and what delta counts as signal. Use when the user says "A/B test this", "write variants", "give me options", "test different angles", "copy variants", "ab copy", "test this headline", "multiple versions", "which version is better", or wants to test copy before publishing. For reviewing copy that already exists, see copy-review. For sample size and stopping rules, see growth-experiment. For diagnosing a page rather than its words, see page-cro. For full paid ad sets by platform, see ad-campaign-writer.

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Write measurable acceptance criteria for requirements or user stories. Use when the user says "write acceptance criteria", "definition of done", "how do we test this", "fit criteria", "given when then", "Gherkin scenarios", "what does done look like", "testable conditions", "how will we know this works" - even if they don't explicitly say "acceptance criteria".

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Build a strategic account plan for winning, retaining, or expanding a named account. Use when the user says "build an account plan", "strategic account planning", "how do I grow this account", "account expansion strategy", "whitespace analysis", "QBR prep", "key account review", "map the stakeholders in this account", "land and expand plan", "how do I break into [company]", or wants a structured strategy for a specific account.

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Write paid ad copy for LinkedIn, Google, Meta, and YouTube. Produces multiple variants per platform, each targeting a different Eugene Schwartz awareness level and Cialdini persuasion principle. Use when the user asks for ad copy, ad variants, "write LinkedIn ads", Google ads, Meta ads, paid social copy, ad headlines, or wants creative for a paid campaign. Reads brand voice, ICP, and positioning from knowledge/. For organic social posts, see linkedin-post or social-calendar. For the landing page the ad points at, see landing-page-writer. For testing variants, see ab-copy-writer.

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

Design clean and consistent APIs. Use when the user says "design an API", "API design review", "design these endpoints", "REST API for X", "GraphQL schema", "API contract", "how should this endpoint work", "review my API design", or wants to create or improve an API interface - even if they don't explicitly say "API design".

日本語の概要は準備中です。原文の説明を表示しています。

qa-aman/claude-skills202026年9月10日 更新

qa-aman のスキルをすべて見る

このスキルの問題を報告する