Team lead
π§ Agent skills for founders who code, parent, and build companies. Productivity systems, deep work, baby sleep guides, burnout recovery β all evidence-based. For OpenClaw, Claude Code, Codex CLI & Cursor.
npx -y skills add sanada123/openclaw-skills --skill team-leadAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 2 stars2 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Engineering management for first-time and experienced tech leads. Covers: 1:1 meetings, hiring engineers, interview frameworks, giving feedback (radical candor), sprint planning, retrospectives, standups, managing underperformers, delegation, first 90 days as manager, remote team management, performance reviews, team culture, technical decision-making, and growing engineers. Not HR theory β practical playbooks from real engineering orgs. Use when discussing team management, hiring, 1:1s, feedback, sprints, retros, performance, delegation, or engineering leadership. Triggers on: "team lead", "engineering manager", "1:1", "hire", "interview", "feedback", "sprint", "retro", "performance review", "underperformer", "delegation", "remote team", Χ ΧΧΧΧ, Χ¦ΧΧΧͺ, ΧΧΧΧ‘, Χ¨ΧΧΧΧ, Χ€ΧΧΧΧ§.
The file declares its own license as MIT. That is the authorβs claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
14.8 KB, as published. Nobody here has run it
Team Lead
The engineering management playbook for people who'd rather write code.
You got promoted (or promoted yourself) because you're a great engineer. Nobody taught you how to manage humans. This skill fills that gap β practical frameworks that work in real engineering teams, from 2-person startups to 50-person departments.
Core Philosophy
PEOPLE > PROCESS. A great team with bad process will fix the process. Bad team with great process will fail.
CLARITY > CONSENSUS. Decide, communicate, move. Don't committee-ize everything.
FEEDBACK > SILENCE. Uncomfortable truth now beats explosive problem later.
TRUST > CONTROL. Hire well, set expectations, then get out of the way.
GROW > RETAIN. People leave managers, not companies. Grow them or lose them.
When to Activate
- Running or joining a team as lead/manager
- Preparing for 1:1 meetings
- Hiring: writing job descriptions, interviewing, evaluating
- Giving difficult feedback
- Sprint planning, retros, standups
- Dealing with underperformers
- Remote team management
- First-time manager seeking guidance
- Performance review season
1. The 1:1 β Your Most Important Meeting
Structure (30 min, weekly, never cancel)
First 10 min: THEIR agenda
"What's on your mind?"
"What do you need from me?"
"Anything blocking you?"
Middle 10 min: YOUR agenda
Progress on goals
Feedback (both ways)
Career development check-in
Last 10 min: ACTION ITEMS
What will they do before next 1:1?
What will YOU do before next 1:1?
Write it down. Both of you.
Questions That Actually Work
| Purpose | Question |
|---|---|
| Open up | "What's the thing you're spending the most energy on right now?" |
| Unblock | "If you could change one thing about how we work, what would it be?" |
| Career | "What skill do you want to be known for in 2 years?" |
| Feedback | "What's one thing I could do differently to help you more?" |
| Satisfaction | "On a scale of 1-10, how happy are you at work? Why not higher?" |
| Trust | "Is there anything you're hesitant to tell me?" |
1:1 Anti-Patterns
| β Don't | β Do |
|---|---|
| Status update meeting | Coaching + unblocking session |
| You talking 80% of the time | They talk 70%+ |
| Cancel when busy | Reschedule, never cancel |
| Same questions every week | Rotate, go deeper |
| Skip when "everything is fine" | "Fine" often means "I'm not telling you" |
2. Hiring Engineers
Job Description Formula
[Company β 1 line what you do]
We're looking for [role] to [impact in 1 sentence].
## What you'll do
- [3-5 concrete responsibilities, not vague buzzwords]
## What you bring
- [2-3 must-haves (be honest β only REAL requirements)]
- [2-3 nice-to-haves (explicitly labeled)]
## What we offer
- [Salary range (post it β saves everyone time)]
- [2-3 actual perks, not "fast-paced environment"]
## How to apply
[Clear instructions. 1 step, not 5.]
Interview Framework (4 stages)
| Stage | Duration | What to Assess | Who |
|---|---|---|---|
| 1. Screen | 30 min | Communication, motivation, basic fit | Recruiter or you |
| 2. Technical | 60-90 min | Coding ability, problem-solving, technical depth | Senior engineer |
| 3. System Design | 45-60 min | Architecture thinking, trade-offs, real-world experience | You or CTO |
| 4. Culture/Values | 30-45 min | Collaboration, ownership, growth mindset, team fit | Team member |
Interview Scorecard
Rate each 1-5:
TECHNICAL
[ ] Problem-solving approach (not just answer)
[ ] Code quality and readability
[ ] System design and trade-offs
[ ] Depth in their claimed expertise
COLLABORATION
[ ] Communication clarity
[ ] How they handle disagreement
[ ] Do they ask good questions?
[ ] Would the team enjoy working with them?
GROWTH
[ ] Self-awareness of weaknesses
[ ] Learning from past mistakes
[ ] Curiosity / asks questions back
[ ] Takes ownership vs blames others
DECISION: Strong Hire / Hire / No Hire / Strong No Hire
Hiring Red Flags
| Red Flag | What It Signals |
|---|---|
| Can't explain a past project simply | Low communication skills |
| "That was someone else's fault" consistently | No ownership |
| No questions about the team or product | Not genuinely interested |
| Perfect answers, no depth when probed | Rehearsed, not experienced |
| Talks badly about previous employers | Will talk badly about you |
| Salary is the ONLY consideration | Mercenary β will leave for $5K more |
3. Giving Feedback (Radical Candor)
The Framework
CARE PERSONALLY
|
Ruinous | Radical
Empathy | Candor βββ THE GOAL
|
ββββββββββββββββββΌβββββββββββββ CHALLENGE DIRECTLY
|
Manipulative| Obnoxious
Insincerity | Aggression
|
Radical Candor = Care Personally + Challenge Directly. Both. Always.
Feedback Formula: SBI (Situation, Behavior, Impact)
"In [SITUATION], when you [BEHAVIOR], the impact was [IMPACT]."
Example (corrective):
"In yesterday's code review [situation], when you dismissed Sarah's
suggestion without explaining why [behavior], she felt her input
wasn't valued and stopped contributing [impact]."
Example (positive):
"In the client demo this morning [situation], when you handled
that unexpected question with a clear, honest answer [behavior],
the client's confidence in our team visibly increased [impact]."
Feedback Timing
| When | How |
|---|---|
| Positive feedback | Public (in Slack, standup) or private. Both work. |
| Corrective feedback | ALWAYS private. Never in front of others. |
| Timing | Within 48 hours. Stale feedback loses power. |
| Frequency | Weekly minimum. Don't save it all for reviews. |
Having Difficult Conversations
1. STATE the issue clearly (SBI format)
2. PAUSE β let them respond
3. LISTEN β really listen, don't plan your rebuttal
4. AGREE on the problem (or understand their perspective)
5. CO-CREATE a plan to fix it
6. SET a follow-up date
7. DOCUMENT what was agreed
4. Sprint Planning & Agile (Practical)
Standup That Doesn't Suck (10 min max)
Each person, 2 minutes:
1. What did I ship yesterday? (not "worked on" β "shipped")
2. What will I ship today?
3. Am I blocked? (if yes, who can unblock me?)
Rules:
- Stand up (it really helps keep it short)
- No problem-solving during standup β "let's take that offline"
- Start on time even if people are missing
- If remote: cameras on, async updates OK for simple days
Sprint Planning (2 hours max for 2-week sprint)
1. Review: what shipped last sprint? What didn't? (15 min)
2. Priorities: what MUST ship this sprint? (from PM/product) (15 min)
3. Breakdown: split each priority into tasks β€ 1 day (45 min)
4. Estimate: rough T-shirt sizes β S (half day), M (1 day), L (2-3 days) (15 min)
5. Commit: team agrees on sprint scope (15 min)
6. Buffer: leave 20% capacity empty for bugs/emergencies (always)
Retrospective Template (45 min, end of sprint)
1. WHAT WENT WELL? (10 min)
Each person: 1-2 things that worked
2. WHAT DIDN'T GO WELL? (10 min)
Each person: 1-2 pain points (no blame)
3. ROOT CAUSE for top 2 issues (10 min)
"Why did this happen?" (ask 5 times β 5 Whys technique)
4. ACTION ITEMS (10 min)
For each issue: WHO will do WHAT by WHEN
Max 3 action items per retro (more = none get done)
5. CHECK last retro's action items (5 min)
Did we actually follow through?
5. Managing Underperformers
The 3-Strike Framework
Strike 1: Feedback conversation
- Clear SBI feedback
- Ask: "What's going on?" (sometimes there's a personal issue)
- Co-create improvement plan with specific, measurable goals
- Timeline: 2-4 weeks
- Document the conversation
Strike 2: Formal warning
- Progress check: did they improve?
- If not: escalate to formal PIP (Performance Improvement Plan)
- Specific goals, weekly check-ins, 30-day timeline
- Document everything in writing
- Involve HR if you have one
Strike 3: Exit
- If no improvement after PIP β it's time to part ways
- Do it respectfully: "This isn't working for either of us"
- Provide notice, help with transition
- Don't drag it out β it hurts everyone, including them
Common Causes (and actual fixes)
| Symptom | Possible Cause | Fix |
|---|---|---|
| Missing deadlines | Unclear expectations | Be more specific about what "done" means |
| Low quality code | No code review culture | Implement PR reviews, pair programming |
| Not participating | Feeling unheard | Ask them directly in 1:1. Make space. |
| Constant conflicts | Cultural mismatch | Mediate once. If pattern continues β exit. |
| Good work, bad attitude | Burnout or personal issue | Have an honest conversation. Offer support. |
6. Delegation
The Delegation Matrix
| Task | Urgency | Who Should Do It |
|---|---|---|
| Critical + only you can do | High | You β now |
| Critical + someone else can do | High | Delegate with clear deadline + check-in |
| Important + develops someone | Medium | Delegate as growth opportunity |
| Routine | Low | Delegate permanently β this isn't your job anymore |
| Shouldn't be done at all | β | Delete it |
How to Delegate Well
1. WHAT: Clear deliverable ("deploy the new API endpoint", not "work on the API")
2. WHY: Context ("this unblocks the mobile team's sprint")
3. WHEN: Deadline ("by Thursday EOD")
4. HOW MUCH AUTONOMY:
- Level 1: "Do exactly this" (junior)
- Level 2: "Research options and recommend" (mid)
- Level 3: "Decide and tell me what you decided" (senior)
- Level 4: "Decide and do it" (trusted senior)
5. CHECK-IN: "Let's sync on Wednesday to see how it's going"
Delegation Anti-Pattern: The Boomerang
If every delegated task comes back to you β you're not delegating, you're just creating extra steps.
Fix: Give more context upfront. Accept 80% quality. Let them fail (small stakes) and learn.
7. Remote Team Management
The Remote Playbook
| Challenge | Solution |
|---|---|
| Isolation | Weekly team social (non-work). 15 min. Optional but encouraged. |
| Miscommunication | Write it down. "If it wasn't written, it wasn't said." |
| Time zones | Define overlap hours (4h minimum). Async for everything else. |
| Trust | Focus on output, not online status. Judge by results. |
| Onboarding | Over-document. Assign a buddy. First week = 100% guided. |
| Camera fatigue | Cameras on for 1:1s and team meetings. Off for focus time. |
Async Communication Guidelines
For your team, establish:
1. Response time expectations: Slack = 2h, Email = 24h
2. "Urgent" channel for real emergencies only
3. Meeting-free blocks (e.g., no meetings before 11 AM)
4. Friday = low-meeting day
5. Document decisions in shared doc, not chat (chat is ephemeral)
8. First 90 Days as New Manager
Day 1-30: Listen
- 1:1 with every team member (learn their goals, frustrations, strengths)
- 1:1 with your manager (understand expectations)
- 1:1 with adjacent team leads (understand dependencies)
- Read all existing docs (architecture, runbooks, retro notes)
- Don't change anything yet. Earn trust first.
Day 31-60: Diagnose
- Identify the top 3 problems (team will have told you by now)
- Pick ONE to fix (not three β one)
- Propose your plan to the team (not decree β propose)
- Start weekly 1:1 rhythm if not already happening
- Begin giving feedback (both positive and constructive)
Day 61-90: Act
- Ship the ONE fix you proposed
- Measure: did it actually improve things?
- Start on problem #2
- By now: team trusts you, you understand the system, you've earned credibility
The #1 mistake: Changing everything in week 1. You don't have context yet. Listen first.
9. Performance Reviews
Review Template (Quarterly or Semi-Annual)
## Performance Review: [Name] β [Period]
### Summary (2-3 sentences)
[Overall assessment: exceeding / meeting / below expectations]
### Key Accomplishments
1. [Shipped X, resulting in Y impact]
2. [Led Z initiative]
3. [Grew in area W]
### Areas for Growth
1. [Specific skill/behavior to improve]
2. [With concrete next steps]
### Goals for Next Period
1. [Measurable goal #1]
2. [Measurable goal #2]
3. [Development goal β skill to build]
### Compensation (if applicable)
[Raise / promotion / no change β with clear reasoning]
The No-Surprises Rule
Nothing in a performance review should be a surprise. If you've been doing weekly 1:1s with regular feedback, the review is just a summary of conversations already had.
If someone IS surprised β you failed at giving ongoing feedback, not them.
Output Standards
When giving management advice:
- Be practical: Templates and scripts, not theory
- Be direct: "Fire them" when that's the answer, not "consider realignment"
- Acknowledge discomfort: Management is emotionally hard. It's OK to feel uncomfortable.
- Context-aware: Startup with 3 people β scale-up with 50
- Reference frameworks: SBI, Radical Candor, RACI β but explain, don't just name-drop
- Defend IC time: Managers who never code lose technical credibility. Protect some maker time.
Because the hardest bugs to fix aren't in the code β they're in the team.
Contributing: Engineering managers, tech leads, CTOs with battle scars β your playbooks are welcome. Open an issue or PR at github.com/sanada123/openclaw-skills.