feedback-writing
Write constructive feedback using Situation-Behavior-Impact for performance, peer and 360 reviews.
Skill by poorvith-mp
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/poorvith-mp-feedback-writing
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: writing
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/feedback-writing
What's inside
- Skill name
feedback-writing- Description
- Write constructive feedback using Situation-Behavior-Impact for performance, peer and 360 reviews. Use when drafting constructive performance reviews, peer praise, or coaching.
- License
- MIT
- Instructions
- 88 lines in SKILL.md · view on GitHub
Files (3)
- SKILL.md
- routing.yaml · evals/
- feedback-writing.skill
SKILL.md (first lines)
# feedback-writing
## Core Philosophy
Constructive feedback fails when it is delivered as subjective character criticism or wrapped in cowardly "feedback sandwiches" (praise-criticism-praise) that confuse the recipient. High-impact professional feedback is grounded in objective behavioral observation, clear business impact, and psychological safety. Utilizing the Center for Creative Leadership’s **Situation-Behavior-Impact (SBI)** model, feedback separates observable actions from assumed motives and establishes clear forward-looking behavioral commitments.
---
## 4-Step Constructive Feedback Framework (The SBI Model)
### Step 1: Situation (Anchor in Exact Time & Place)
1. **Specific Temporal Anchoring**:
- Establish the precise context so the recipient's memory is grounded:
- *Bad*: "You're often aggressive in team meetings."
- *Good (SBI Situation)*: "Yesterday during the 2:00 PM architecture review for the billing migration..."
### Step 2: Behavior (Observable, Video-Camera Facts)
1. **The "Video-Camera" Test**:
- Describe only what a video camera could record. Eliminate adjectives, personality traits, and mind-reading assumptions:
- *Bad*: "You were rude and dismissive to the junior engineer."
- *Good (SBI Behavior)*: "When Sarah was presenting her database schema proposal, you interrupted her three times and said 'this approach is completely stupid' before she finished explaining."
### Step 3: Impact (The Measurable Outcome on Team & Business)
1. **Explaining the Repercussions**:
- Articulate the emotional, team, or operational consequences:
- *SBI Impact*: "Sarah stopped sharing her ideas for the rest of the meeting, the junior team members became visibly hesitant to ask technical questions, and we missed exploring an edge case that delayed our sprint sign-off."
### Step 4: Alternative Behavior & Collaborative Commitment (SBI-A)
1. **Forward-Looking Alignment**:
- Agree on concrete future behavior rather than litigating the past:
- *"In future architecture reviews, I need you to let the presenter finish their slide deck before critiquing, and phrase disagreements around technical trade-offs rather than labels. How does that sound to you?"*
---
## Deliverable Format: Performance Feedback Record (`FEEDBACK-MEMO.md`)
```markdown
# Constructive Feedback Memo: [Recipient Name]
*Date: [YYYY-MM-DD] | Manager / Reviewer: [Name, Title]*
## 1. Feedback Type & Context
- **Type**: [Developmental / Performance Coaching / Peer 360 Review]
- **Setting**: 1-on-1 private sync
## 2. SBI Feedback Framework
- **Situation**:
During yesterday's sprint planning session at 10:00 AM...
- **Behavior (Observable Facts)**:
You committed to delivering the authentication refactor by Wednesday, but did not update the Jira ticket or notify the team until Friday afternoon when CI builds failed.
- **Impact (Business & Team Consequence)**:
Frontend engineers were blocked from integrating the new login flow for two days, and we had to reschedule the client demo from Thursday to next week.
## 3. Required Future Behavior (The Commitment)
- Update ticket status within 2 hours if an unexpected blocker delays a committed deadline.
- Post an async update in `#eng-blockers` as soon as an issue is identified so teammates can swarm.
## 4. Recipient Response & Agreed Next Action
- [Notes on recipient's perspective and confirmed commitment].
- Check-in review date scheduled for: [Date, 2 weeks out].
```
---
## Worked Example: Senior Engineer Code Review Feedback
- **Situation**: During pull request reviews on Tuesday.
- **Behavior**: Left 18 one-line comments saying "rewrite this" without explanations or suggestions.
- **Impact**: Junior developer felt demoralized and spent 6 hours guessing what was wrong rather than learning the team standard.
- **Resolution**: Aligned on using conventional review prefixes (`nit:`, `blocking:`) with concrete code examples. Review turnaround improved 40%.
---
## Verification Checklist
- [ ] Situation specifies exact time, meeting, or event.
- [ ] Behavior describes observable actions that pass the "video-camera test".
- [ ] Impact clearly explains business, team, or project consequences.
- [ ] Zero subjective character assassinations ("unprofessional", "lazy", "arrogant").
- [ ] Concludes with a concrete forward-looking behavioral agreement.
---
## Anti-Patterns
- **The Feedback Sandwich**: Hiding critical performance issues between fake compliments, leaving the recipient thinking everything is great.
- **Mind Reading**: Claiming to know someone's intent ("You obviously don't care about this company").
- **Public Shaming**: Delivering developmental criticism in a public Slack channel or team meeting instead of 1-on-1.From poorvith-mp/skills-writing, MIT.
More from poorvith-mp/skills-writing
- 0.452score
book-writing
poorvith-mp
Develop book-length work: premise, chapter architecture, narrative voice, and full revision.
SkillAFreshness: AClaude Code +9 - 0.452score
executive-summary
poorvith-mp
Compress complex input into one page: recommendation first, then the supporting case.
SkillAFreshness: AClaude Code +9 - 0.452score
grant-writing
poorvith-mp
Write grant applications for nonprofits, research institutions and social enterprises, including prospect research.
SkillAFreshness: AClaude Code +9 - 0.452score
historical-accuracy
poorvith-mp
Check period detail, material culture and historiography for anachronism.
SkillAFreshness: AClaude Code +9 - 0.452score
job-descriptions
poorvith-mp
Write role-accurate JDs with responsibilities, required versus preferred qualifications and pay bands.
SkillAFreshness: AClaude Code +9 - 0.452score
line-editing
poorvith-mp
Tighten a draft: cut filler, vary rhythm, sharpen verbs, and make every sentence earn its place.
SkillAFreshness: AClaude Code +9 - 0.452score
meeting-notes
poorvith-mp
Turn notes or a transcript into decisions, action items with owners and deadlines, and a parking lot.
SkillAFreshness: AClaude Code +9 - 0.452score
sop-writing
poorvith-mp
Write procedures anyone can follow: numbered steps, owners, inputs, outputs, escalation and executable templates.
SkillAFreshness: AClaude Code +9 - 0.452score
talk-writing
poorvith-mp
Write a conference talk or keynote: through-line, one idea per slide, opening hook and honest timing.
SkillAFreshness: AClaude Code +9 - 0.452score
technical-docs
poorvith-mp
Write API references, user guides, setup manuals and architecture docs with consistent terminology.
SkillAFreshness: AClaude Code +9 - 0.452score
thread-to-article
poorvith-mp
Expand an X or LinkedIn thread into a structured article with added context, headings and links.
SkillAFreshness: AClaude Code +9 - 0.452score
translation
poorvith-mp
Translate while preserving register, idiom and domain terms, flagging what doesn't carry across.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 2
- Open issues
- 0
- Last commit (this folder)
- Sep 12, 2026 (1 month ago)
- Last release
- Oct 2, 2026 (26 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/poorvith-mp/skills-writing
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from poorvith-mp
- 0.495score
blender-modeling
poorvith-mp
Model, retopologise, UV unwrap and bake game-ready assets with correct scale, pivots and polycount.
SkillAFreshness: A★ 5Claude Code +9 - 0.495score
game-audio
poorvith-mp
Integrate FMOD and Wwise, build adaptive music systems, spatial audio and audio performance budgets.
SkillAFreshness: A★ 5Claude Code +9 - 0.495score
anti-cheat
poorvith-mp
Enforce server authority and defend against speed hacks, memory editing, and exploits.
SkillAFreshness: A★ 5Claude Code +9 - 0.495score
blender-animation
poorvith-mp
Rig, skin and animate characters and props, and export clean animation sets to an engine.
SkillAFreshness: A★ 5Claude Code +9 - 0.495score
game-ai
poorvith-mp
Design NPC behaviour: behaviour trees, state machines, pathfinding, perception and difficulty tuning.
SkillAFreshness: A★ 5Claude Code +9 - 0.495score
game-build-pipeline
poorvith-mp
Automate platform builds, asset cooking, console certification prep, and Steam or itch.io depot publishing.
SkillAFreshness: A★ 5Claude Code +9
Cite this page
feedback-writing skill. Agent Mart. https://agentmart.dev/skills/poorvith-mp-feedback-writing (last updated 2026-10-24)