borghei/Claude-Skills / project-management/career/pm-1on1s
pm-1on1s
VerifiedBuilds 1:1 agendas and kickoff scripts for a PM's manager, EM, designer, reports and cross-functional partners. Use when starting a new 1:1, a 1:1 has turned into status updates, or you need to give a partner hard feedback.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.6 KB |
| assets/1on1_notes_template.md | markdown | 2.0 KB |
| assets/kickoff_template.md | markdown | 3.3 KB |
| assets/manager_1on1_agenda.md | markdown | 3.0 KB |
| examples/sr-pm-first-month-with-em-partner.md | markdown | 12.4 KB |
| references/1on1-playbook.md | markdown | 7.9 KB |
| references/one-on-one-templates.md | markdown | 10.2 KB |
| references/red-flags.md | markdown | 12.1 KB |
references/red-flags.md
View on GitHub# Red Flags: PM 1:1s
> Common ways this skill's output goes wrong — concrete examples, why they're bad, and how to fix them. Pair with the SKILL.md and Troubleshooting table.
## How to use this document
When you have just drafted a 1:1 agenda, written notes, or designed a question bank, scan the red flags below before your next 1:1. Each red flag shows a *bad* and *good* version of the same artifact or interaction.
---
## Red Flag 1: Status Update Theater
**Symptom.** The 1:1 with your manager is 25 minutes of "here is what I worked on this week" with no questions, no asks, no growth conversation.
**Why it's bad.** Status updates belong in async writeups (Slack, weekly digest, project tracker). The 1:1 is your highest-bandwidth conversation with your manager; spending it on read-aloud status burns the slot. Manager Tools' framework explicitly puts status at the *bottom* of the agenda.
**Bad example:**
> Agenda: "1. Project A update. 2. Project B update. 3. Project C update. 4. Any questions?"
**Good example:**
> "Async sent Monday: 5-bullet status. 1:1 agenda: (a) 1 ask for help (Project B stakeholder blocked), (b) 1 growth topic (preparing for next-level promo discussion), (c) feedback I owe — both directions, (d) what is on your mind that I should know about."
**How to catch it.** If 60%+ of the 1:1 is updating your manager on things they could read async, you have status theater.
---
## Red Flag 2: No Trust-Building or Personal Connection
**Symptom.** Every 1:1 starts with "okay, first item..." and ends on a status note. No check-in on the human at the other end of the table.
**Why it's bad.** Kim Scott's Radical Candor frame requires *caring personally* before *challenging directly*. A 1:1 with no human dimension cannot deliver hard feedback; the trust account is empty when you need to make a withdrawal.
**Bad example:**
> "Hi. Item 1: roadmap. The sprint is at risk because..."
**Good example:**
> "Hi, before we get into it — how's your week been? You mentioned your kid's exam last time, how did that go?" (2-3 minutes, sincere.)
**How to catch it.** Do you know one non-work fact about the person at the other end of each recurring 1:1? If not, your 1:1s are transactional.
---
## Red Flag 3: One-Directional Feedback Flow (Only Down)
**Symptom.** Your manager gives you feedback; you never give feedback to your manager. Or you give feedback to your reports but never invite it back.
**Why it's bad.** Feedback that only flows downhill is performance management, not coaching. Manager Tools and Radical Candor both require *upward* feedback explicitly. Skipping it leaves your manager unable to improve and signals that feedback is a sword, not a tool.
**Bad example:**
> Manager: "Here's what you could improve." You: "Thanks, I'll work on that."
**Good example:**
> Manager: "Here's what you could improve." You: "Thanks, I'll work on that. One piece of feedback for you: when we discussed the roadmap last week, I felt the priorities shifted faster than I could communicate to the team. Could we agree on a 'no changes within sprint' rule?"
**How to catch it.** In your last 4 1:1s, have you delivered at least 1 piece of upward feedback? If zero, the feedback channel is one-way.
---
## Red Flag 4: GROW Model Skipped — Jumping to Will (Solutions)
**Symptom.** Report or partner brings up an issue; you immediately offer a solution. Goal/Reality/Options are skipped.
**Why it's bad.** GROW (Whitmore) is in this skill because the *value* of a coaching conversation is helping the other person think, not solving their problem. Skipping to Will is faster but produces dependency and underdeveloped reports.
**Bad example:**
> Report: "I'm struggling with the engineering team's velocity." You: "You should set a clearer DoD and run a velocity retro."
**Good example:**
> Report: "I'm struggling with the engineering team's velocity." You: "What does 'good' look like for you here? [Goal] What's actually happening today? [Reality] What options have you considered? [Options] Which of those will you try first? [Will]"
**How to catch it.** If you find yourself saying "you should..." in the first 5 minutes of a topic, you skipped GROW.
---
## Red Flag 5: Skip-Level Information Leakage
**Symptom.** You repeat to your skip-level (or to your manager's manager) something a report told you in confidence in a 1:1.
**Why it's bad.** 1:1 confidentiality is the foundation of psychological safety. One breach kills the channel — the report stops sharing real concerns, retreats to status theater, and looks for a new manager or a new job.
**Bad example:**
> In skip-level: "Jamie mentioned in our 1:1 that she's frustrated with the EM."
**Good example:**
> In skip-level: "A common theme in my 1:1s is that PMs feel unclear on engineering priorities. I'm working with my reports to surface specifics; would you support a cross-functional alignment session?"
**How to catch it.** Before mentioning anything from a 1:1 in any other meeting, ask: "did this person give me permission to share this?" If not, abstract or omit.
---
## Red Flag 6: No Notes, No Follow-Through
**Symptom.** You take no notes. Two weeks later, you cannot remember what you agreed. The same topics resurface; commitments evaporate.
**Why it's bad.** A 1:1 without notes is improv. The Manager Tools framework requires written follow-ups precisely because human memory is unreliable and trust is built by following through. Reports notice when their issues vanish into the ether.
**Bad example:**
> No shared document. Both parties go to next 1:1 trying to remember what they agreed.
**Good example:**
> Shared rolling doc. Each 1:1 has dated agenda, decisions, and action items with owner. Items not closed last time are carried forward. 15-minute pre-read before each 1:1.
**How to catch it.** Can you produce a 4-month history of agreed actions per 1:1 relationship? If not, your follow-through is fictional.
---
## Red Flag 7: Skipping or Constantly Rescheduling 1:1s
**Symptom.** "Crazy week, let's skip the 1:1." Or 1:1s get moved 3 times then dropped.
**Why it's bad.** Cadence is the point. The 1:1 is the *only* guaranteed conversation between two people in the week. Skipping it during high-stress weeks is exactly when it would have caught problems early. A skipped 1:1 signals "you are not important enough" — even when meant otherwise.
**Bad example:**
> Slack: "Crazy launch week, let's skip our 1:1 and catch up next week."
**Good example:**
> Slack: "Crazy launch week. Shrinking 1:1 to 15 min and tightening agenda to just blockers and one ask. Still meeting at 2pm."
**How to catch it.** Over the last 8 scheduled 1:1s, what % actually happened? If under 75%, cadence is broken.
---
## Red Flag 8: Coaching Without Permission (Unsolicited)
**Symptom.** Partner brings up a problem; you launch into coaching mode without asking if they want coaching or just to vent.
**Why it's bad.** Unsolicited coaching feels patronizing. Sometimes people need to be heard, not coached. Asking permission — "would it be more helpful for me to listen or to ask coaching questions?" — preserves agency.
**Bad example:**
> Partner: "The launch is a disaster, I'm furious." You: "What does great look like here? What have you tried?"
**Good example:**
> Partner: "The launch is a disaster, I'm furious." You: "Sounds rough. Do you want to vent for a few, want me to help think it through, or want a recommendation?"
**How to catch it.** In your last 5 coaching attempts, did you ask permission first? If never, you are imposing the frame.
---
## Red Flag 9: Treating Direct Reports' 1:1s as Performance Reviews
**Symptom.** Every 1:1 with a report becomes a mini-review: "here's what I noticed you doing wrong this week."
**Why it's bad.** 1:1s are *their* time, not yours. Performance feedback belongs in a separate, regular feedback cadence (Radical Candor's "in the moment" or scheduled performance check-ins). A 1:1 that is a feedback-delivery slot is dreaded and skipped.
**Bad example:**
> Manager-to-report 1:1: 80% manager speaking, 20% report. Most of manager talk is corrective feedback.
**Good example:**
> Manager-to-report 1:1: 70% report speaking on their topics, 20% growth/career, 10% feedback (both directions). Corrective feedback delivered in-the-moment, not stockpiled.
**How to catch it.** In the last 1:1, what % of speaking time was theirs vs yours? Under 50% theirs is a manager-speech red flag.
---
## Red Flag 10: No Growth or Career Conversation Ever
**Symptom.** 1:1s cover only current work. Growth, promotion, skill development, and longer-arc career topics never come up.
**Why it's bad.** The PM career ladder is steep and signals are noisy. Reports who never discuss growth assume they are not on track for promotion; partners who never discuss skills stagnate. Manager Tools recommends a *quarterly* growth-focused 1:1 minimum.
**Bad example:**
> 26 weeks of 1:1s. Zero mentions of promotion, skill gaps, or career arc.
**Good example:**
> Standing quarterly agenda: "Where do you want to be in 12 months? What's the next-level rubric saying? What experiments are you running this quarter? How can I sponsor your growth?"
**How to catch it.** Search your 1:1 notes for "growth", "career", "promotion", "next level" over the last 90 days. If zero hits, growth is invisible.
---
## Red Flag 11: Using 1:1s for Tactical Decisions That Should Be Group Meetings
**Symptom.** Strategic or cross-functional decisions get made in 1:1s, then the team is informed after the fact.
**Why it's bad.** Decisions made in 1:1s exclude the people who should have weighed in. The decision is fragile (no group buy-in) and the people who should have heard the reasoning hear only the outcome.
**Bad example:**
> Manager and PM decide in 1:1 to change the roadmap priorities. Team finds out via revised Jira board.
**Good example:**
> In 1:1: "I think we should re-prioritize, here's why. Can we put this on the next product trio meeting agenda so design and engineering weigh in?"
**How to catch it.** If a decision in a 1:1 affects 3+ people who were not in the 1:1, the decision should move to a group forum.
---
## Red Flag 12: Cross-Functional 1:1 Becomes a Sales Pitch
**Symptom.** 1:1 with the EM, design partner, or cross-functional lead turns into you trying to sell them on your roadmap.
**Why it's bad.** Partners notice when a 1:1 is one-sided. The relationship is a deposit account; selling without asking burns the deposit. The partner-type-specific templates in this skill exist to balance the conversation across both sides' interests.
**Bad example:**
> 1:1 with EM: PM spends 40 minutes pitching why mobile work should be prioritized. EM nods, leaves drained.
**Good example:**
> 1:1 with EM: 5 min context-sharing, 15 min listening (what are EM's pressures, team capacity reality), 10 min joint priorities discussion, 10 min process improvements both can make.
**How to catch it.** In your last cross-functional 1:1, did you ask 5+ questions about *their* world? If not, it was a pitch.
---
## Red Flag Quick Reference
| # | Anti-pattern | One-line check |
|---|---|---|
| 1 | Status Update Theater | Status moved to async; 1:1 is growth/asks/feedback? |
| 2 | No Trust-Building | Do you know one non-work fact about them? |
| 3 | One-Directional Feedback | Have you given upward feedback in last 4 1:1s? |
| 4 | Skipping to Solutions | Did you ask Goal/Reality/Options before Will? |
| 5 | Skip-Level Information Leakage | Permission to share before mentioning elsewhere? |
| 6 | No Notes / No Follow-Through | Can you produce 4 months of dated actions? |
| 7 | Skipping or Rescheduling | Last 8 1:1s, % actually happened above 75%? |
| 8 | Coaching Without Permission | Did you ask listen/coach/recommend? |
| 9 | 1:1 as Performance Review | Report speaks 50%+ of the time? |
| 10 | No Growth Conversation | Growth topic in last 90 days of 1:1 notes? |
| 11 | Group Decisions in 1:1 | Decisions affecting 3+ moved to a group? |
| 12 | Cross-Functional 1:1 as Pitch | 5+ questions about their world? |
## Related Reading
- SKILL.md Troubleshooting section (for symptom -> root cause -> resolution)
- references/templates/ (for partner-type-specific 1:1 templates, if present)
- pm-onboarding/references/red-flags.md (for new-relationship 1:1 patterns)
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| brainstormingobra/superpowersYou MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation). | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| brand-guidelinesanthropics/skillsApplies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| frontend-designanthropics/skillsGuidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| mcp-builderanthropics/skillsGuide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |