feedback
Skill114 starsChanged 7 months agoMITView on GitHub
Covers both giving and getting feedback — structures and scripts feedback conversations (positive, constructive, or behavioral) and provides techniques for drawing honest feedback from your own team. Produces SBI-framed feedback statements, opening lines for hard conversations, scripts for real situations, ways to handle resistance, and methods for extracting real feedback from reports. Use when the user wants to give someone feedback, says "how do I tell someone," "this person is struggling," "address a behavior," "hard conversation," "someone is underperforming," "praise this person," "write feedback for," "I need to say something," "difficult conversation," "get feedback from my team," "my team won't give me feedback," "blind spots," or "what does my team think of me." Do NOT use for formal annual or performance reviews (use performance-reviews) or sensitive HR situations that go beyond feedback (use difficult-situations).
26 other projects use a version of this.
Adopted by
Most recently taken up by
- grainulation/grainulatortheir own version, first seen Nov 3, 2026
- menot-you/n-soultheir own version, first seen Nov 1, 2026
- LightconeResearch/agent-skillstheir own version, first seen Nov 1, 2026
- tydev-new/10xjobs-careercoachtheir own version, first seen Nov 1, 2026
- ratomirjovanovic/pandawardtheir own version, first seen Nov 1, 2026
- grace-method/grace-plugintheir own version, first seen Oct 25, 2026
Used with
Projects with it also use setup, status and 6 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
VercelSkillsetup
Set up Vercel CLI and project configuration. Use when the user says "set up Vercel", "configure Vercel", "link to Vercel", or "vercel init".
Also in 5 projects that use this
AnthropicSkillstatus
Case status summary by audience — client-facing (plain language), internal (for the professor), or court-ready (formal caption format per local rules). Same facts, different framing and depth. Use when a student needs to update the client, brief the professor, or prepare a court status report.
Also in 3 projects that use this
GitHubSkillrefactor
Surgical code refactoring to improve maintainability without changing behavior. Covers extracting functions, renaming variables, breaking down god functions, improving type safety, eliminating code smells, and applying design patterns. Less drastic than repo-rebuilder; use for gradual improvements.
Also in 3 projects that use this
AnthropicSkilldebug
Structured debugging session — reproduce, isolate, diagnose, and fix. Trigger with an error message or stack trace, "this works in staging but not prod", "something broke after the deploy", or when behavior diverges from expected and the cause isn't obvious.
Also in 3 projects that use this
BlockSkillplan
Draft a TDD, PRD, or lightweight plan in the conversation without committing it.
Also in 3 projects that use this
SentrySkillreview
Fixture review skill.
Also in 4 projects that use this
Other versions26 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 26
- spotify/portal-ai-plugins★ 2.8ksame name
- suboss87/FDEOps★ 940same name
- davekilleen/Dex★ 509same name
- austintgriffith/ethskills★ 303same name
- grainulation/grainulator★ 86same name
- BaseInfinity/claude-sdlc-harness★ 53same name
- ratomirjovanovic/pandaward★ 50same name
- graphed-inc/marketing-skills★ 30same name
- kennethleungty/claude-music★ 23same name
- coltonjosephdean-rgb/Hyperframes-colton.ai.dean★ 19same name
16 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository28 other files it gives its agents.
- AGENTS.mdAGENTS.md
- 1on1sSkill
- business-literacySkill
- career-developmentSkill
- delegationSkill
- developer-productivitySkill
- difficult-situationsSkill
- em-contextSkill
- em-grid-scorerSkill
- engineer-motivationSkill
- hiringSkill
- influenceSkill
- knowledge-sharingSkill
- management-transitionsSkill
- managing-high-performersSkill
- managing-upSkill
- managing-urgencySkill
- managing-yourselfSkill
- meetingsSkill
- performance-reviewsSkill
- retaining-developersSkill
- roadmap-planningSkill
- shadow-workSkill
- team-compositionSkill
- team-healthSkill
- working-with-architectsSkill
- working-with-pmSkill
- written-communicationSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
18 rules. 0 are common, 12 are distinctive, and 6 are very specific to this project.
0
Common
Most of the 24 edited versions we read keep these.
12
Distinctive
This author's own choices. Worth a look.
6
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Check for EM context first:
- Very specific: line 13
Read .agents/em-context.md if it exists
- Very specific: line 14
If a person is mentioned, look for .agents/reports/[name].md — it may contain their feedback preferences, prior feedback history, and …
- Distinctive: line 15
Use that context — only ask for information not already covered
- Very specific: line 17
If .agents/em-context.md does not exist, ask for a minimal manager profile first and save it before giving detailed advice: role/title, …
- Very specific: line 19
If a specific person is central to the conversation and .agents/reports/[name].md does not exist, ask for a minimal profile for that person …
- Very specific: line 21
If the conversation reveals durable new context later, update .agents/em-context.md or .agents/reports/[name].md automatically.
- Distinctive: line 26
Keep the first answer concise and useful.
- Distinctive: line 30
Ask at most 2-3 targeted questions only if the missing context changes the advice
- Distinctive: line 55
When helping with feedback, produce wording the manager can actually say:
- Distinctive: line 90
Ask before you assume. A simple "How do you prefer to be recognized when you do great work?" in a 1:1 is enough.
- Distinctive: line 96
Not everyone can read a room.
- Distinctive: line 104
When addressing a performance or behavior problem, people typically move through three stages before accepting feedback:
- Distinctive: line 126
Give the criticism directly: No burying. Say the specific thing, the specific event, the specific impact.
- Distinctive: line 127
Invite them to respond: Ask what they think. Let them engage, not just receive.
- Distinctive: line 145
Then say the thing: Don't circle around it. The opening sets the tone.
- Distinctive: line 147
Silence is data: If someone goes quiet after critical feedback, don't rush to fill the space.
- Very specific: line 191
If the user asks where a framework came from, wants to read the original article, or wants more context on any topic in this skill — read …
“Common” is measured: kept by at least 6 in 10 of the 24 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Feedback
Before Starting
Check for EM context first:
- Read
.agents/em-context.mdif it exists - If a person is mentioned, look for
.agents/reports/[name].md— it may contain their feedback preferences, prior feedback history, and working style - Use that context — only ask for information not already covered
If .agents/em-context.md does not exist, ask for a minimal manager profile first and save it before giving detailed advice: role/title, team size, team mission or ownership area, and current challenge or priority.
If a specific person is central to the conversation and .agents/reports/[name].md does not exist, ask for a minimal profile for that person first and save it before giving detailed advice: title/level, tenure, strengths, and current challenge or growth area.
If the conversation reveals durable new context later, update .agents/em-context.md or .agents/reports/[name].md automatically. Save stable facts and patterns, not guesses, transient frustration, or unresolved interpretations.
Response Style
Keep the first answer concise and useful. Do not dump the whole framework unless the user asks for depth.
Default to:
- State the likely diagnosis or recommendation first
- Ask at most 2-3 targeted questions only if the missing context changes the advice
- Give the next concrete action and, when useful, exact wording the manager can use
- Mention the relevant framework briefly, but do not explain every part of it
- Offer a deeper version only after the direct answer
How to Use This Skill
Identify what the user needs before diving into frameworks:
- Giving positive feedback → Go to Precision: No Weasel Words — it applies equally to praise
- Constructive feedback — know the issue but not how to open → Opening a Critical Feedback Conversation, then Instead of the Compliment Sandwich
- Expecting resistance or pushback → Behavior Change Stages first
- Something happened recently and the user wants to address it now → In-the-Moment Feedback (don't save it for the 1:1)
- Building feedback culture across the whole team → Speed Feedback
- Writing feedback for a review or document → Precision section first; then ask for the specific events to document
- Building a lightweight weekly accountability loop with reports → read
references/extended.md— Weekly Called Shots - Getting honest feedback from your own team → Getting Feedback from Your Team
If the user hasn't given you a specific event yet, ask for it. Vague feedback inputs always produce vague feedback outputs.
Default Response Shape
When helping with feedback, produce wording the manager can actually say:
- Intent: what outcome the feedback should create.
- SBI draft: situation, behavior, impact, in concrete language.
- Opening line: a direct but humane start to the conversation.
- Follow-up question: invite their perspective without softening the message.
- Next step: request, agreement, or observation period.
For praise, keep it specific. For corrective feedback, do not hide the message inside a compliment sandwich.
Precision: No Weasel Words
The most common feedback mistake is vagueness. "You're too blunt in meetings" or "your communication needs improvement" — these feel like feedback but give the person nothing to act on. They're also easy to dismiss: "I don't think I'm blunt."
Specific feedback is factual and indisputable.
Weak: "You're too blunt in meetings." Specific: "In the meeting with Christopher, you cut him off when he started suggesting the React-based solution. We missed an opportunity to check more approaches. I also noticed that Christopher stopped contributing to other meetings after that."
The specific version names the event, the person affected, and the downstream consequence. The other person can't dispute the facts — they can only engage with what happened.
Apply this same standard to positive feedback. "You did a great job" is forgettable. "The way you handled the incident on Thursday — staying calm, communicating clearly to stakeholders while the team debugged — that's exactly what we need in situations like that" is memorable and repeatable.
Before giving feedback, do the homework. Identify the specific event, the specific behavior, and the specific impact. Vague feedback isn't kind — it's just unprepared.
Public Praise Isn't Universal
"Praise in public, criticize in private" is standard management advice. The first half isn't always true.
Some engineers genuinely don't want to be called out in front of the team. They prefer private acknowledgment. Surprising them with public praise — a Slack shoutout, a nomination for a recognition program — can create awkwardness instead of motivation.
Ask before you assume. A simple "How do you prefer to be recognized when you do great work?" in a 1:1 is enough.
Translating Social Cues
Not everyone can read a room. If you're good at this, part of your job is translating for those who aren't.
When a team member says something that lands badly and doesn't notice — upset someone, came across as dismissive, answered too abruptly in a meeting — tell them privately. Not as criticism, but as information: "I think that landed differently than you intended. Here's what I noticed." They often don't know. Telling them is the most useful thing you can do.
Behavior Change Stages
When addressing a performance or behavior problem, people typically move through three stages before accepting feedback:
- Ignore — they don't acknowledge the problem exists
- Deny — they actively push back ("That's not true" / "You're wrong")
- Blame others — they redirect responsibility elsewhere
In a single 1:1, you can often move someone through all three stages. The right response at each stage is different:
- At Ignore/Deny: bring specific facts. Don't argue — present evidence calmly.
- At Blame: don't let them escape into it. Acknowledge partial truths, then redirect back to what they can control.
The goal isn't to win the argument — it's to get to a point where the person can own the problem.
Instead of the Compliment Sandwich
Starting with a compliment and ending with one — sandwiching criticism in between — backfires in two ways: the opening compliment feels insincere (they're braced for the hit), and the closing compliment can bury the critique due to recency effects.
Four steps that work better (from Adam Grant, based on research):
- Explain why you're giving the feedback. Signal that you're on their side: "I'm giving you this because I have high expectations and I know you can reach them." It's harder to reject a hard truth from someone who believes in you.
- Take yourself off a pedestal. Reduce the power asymmetry: "I've benefited from people giving me tough feedback — I'm trying to pay that forward." It makes the feedback less threatening.
- Give the criticism directly. No burying. Say the specific thing, the specific event, the specific impact.
- Invite them to respond. Ask what they think. Let them engage, not just receive.
Opening a Critical Feedback Conversation
From The Making of a Manager (Zhuo): the moments right before a difficult feedback conversation feel uncomfortable — for both people. That discomfort is a good sign. It means the conversation is real.
Comfortable feedback conversations often aren't achieving anything. If neither person feels any friction, the feedback is probably not specific enough, not honest enough, or not about anything that actually matters.
What to say at the start:
The opening doesn't need to be elaborate. What it needs to signal is: you're saying this because you care about the person, not because you're judging them.
"I want to share something that's hard for me to say, and I'm saying it because I think it matters for where you're going."
Or simply naming what's happening: "I have some direct feedback for you. It might feel uncomfortable, and I want you to know it's coming from a place of caring about your success here."
Then say the thing. Don't circle around it. The opening sets the tone; the feedback itself has to be clear, specific, and direct. Softening after a strong opener defeats the purpose.
Silence is data. If someone goes quiet after critical feedback, don't rush to fill the space. Let them process. The silence often means they're actually taking it in — which is exactly what you wanted.
In-the-Moment Feedback
Waiting for the next 1:1 to give feedback costs most of its effectiveness. The closer feedback is to the event, the more likely the person can connect it to what they actually did.
Standup feedback — feedback given in the flow of work, at or near the moment it's relevant. Not a big formal conversation; a direct, brief comment. Example: a developer repeatedly delays code reviews for teammates. Raising this in the standup (or right after, in the hallway) is far more effective than bringing it up three days later in a 1:1 where the person may barely remember the specific instance.
The volume button. In-the-moment feedback doesn't need to be loud or intense. Think of it as tuning down the tone and volume — not the substance. Keep the content clear and direct; deliver it calmly and without drama. Brief, respectful, and close to the event beats formal, thorough, and a week later.
Why timing matters. Humans (like any animal) struggle to connect feedback to action when the gap is too long. Waiting days doesn't just reduce effectiveness — it trains the team to expect feedback only in formal settings, which reduces psychological safety in day-to-day work. Google's research on team effectiveness identified psychological safety as the #1 factor — and the team leader is the primary influence on it.
Speed Feedback: Building a Feedback Habit on the Team
Teams that only give feedback in formal reviews are too slow. Speed feedback is a structured activity for building a team-wide feedback habit.
How it works: Team members pair up in a private space. Each person gets a fixed amount of time to give feedback to the other, then they switch. Pairs rotate until everyone has spoken with everyone else.
The goal isn't depth — it's normalization. Early rounds focus on appreciation and celebrating contributions. As the team gets comfortable, they naturally start including improvement suggestions. The activity makes feedback part of the team's rhythm rather than a once-a-year event.
Works best with 5–8 people. Can be run remotely with breakout rooms and a shared board. Does not replace regular 1:1 feedback — it supplements it.
Getting Feedback from Your Team
Bad managers have many problems, but most of them know it. Good managers have a few invisible ones. When you ask for feedback in a performance review, you'll usually get silence — or "everything's great!" There is always something you can improve.
Three methods that get past the surface:
1. Try different angles. Don't accept the first "I can't think of anything." Keep asking from a different direction: "A behavior you didn't like?" "Something specific to your growth or our 1:1s?" "Something the team as a whole could have done better?" Real feedback is often only reachable because the question kept changing.
2. Share your own difficulties. When you share what you struggle with, people become more comfortable criticizing you. It signals that honest assessment is welcome. If your team only finds out about your struggles via a LinkedIn post, the psychological space for that conversation doesn't exist in your 1:1s.
3. Make it about a specific situation. People won't easily criticize you as a person — but they will talk about a project or an event. Instead of "where can I improve?" ask "what could we have done better in project X?" Then push further: "if we had to do the same project again, what would you change?" The second question is where the real feedback lives.
Dive Deeper
If the user asks where a framework came from, wants to read the original article, or wants more context on any topic in this skill — read references/sources.md for the full list of source articles (with links), books, and course material.
Related Skills
1on1s— Delivering feedback in a 1:1 contextperformance-reviews— Formal written feedback at review time
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

