一键导入
121
Prepare for a 1:1 — pull context from goals, meeting notes, Slack, and tasks to generate talking points
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Prepare for a 1:1 — pull context from goals, meeting notes, Slack, and tasks to generate talking points
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Auto-generate or update description, type, topics, aliases, and graph-backed relationship fields for a file.
Build a focused plan for today based on goals, tasks, and recent activity. Reads from the weekly plan if it exists and updates it.
Check system health — metadata gaps, graph integrity, alias sync, and referential integrity.
Weekly review + compound learnings — review progress, surface blockers, distill insights into Memory
Cross-initiative alignment checker — surfaces misalignment, gaps, duplication, dependency risks, and synergy opportunities across active initiatives in the company and your groups roadmap. Complementary to /pulse (broad discovery); /align focuses on whether your initiatives are pulling in the same direction and have synergies with other company groups. Run ad-hoc or before key decisions.
Autonomously optimize any Agent skill by running it repeatedly, scoring outputs against binary evals, mutating the prompt, and keeping improvements. Based on Karpathy's autoresearch methodology. Use when: optimize this skill, improve this skill, run autoeval on, make this skill better, self-improve skill, benchmark skill, eval my skill, run evals on. Outputs: an improved SKILL.md, a results log, and a changelog of every mutation tried.
| name | 121 |
| description | Prepare for a 1:1 — pull context from goals, meeting notes, Slack, and tasks to generate talking points |
Prepare focused talking points for a 1:1 meeting with a specific person.
When the user invokes /121 [person name], run through all steps in sequence.
If no person name is provided, ask: "Who's the 1:1 with?"
Triggers when no Context/121s/[Person Name].md exists. Run this before Step 1 to build context from scratch so talking points are grounded, not generic.
slack_search_users to find their user ID. Then read the DM channel: last 20 messages or past 30 days, whichever is fewer. Extract: topics discussed, open questions, commitments, tone of the relationship.slack_user_id and slack_dm_channel pre-populated from the Slack lookup. Then proceed to Step 1 which will now find the doc.If Slack MCP is unavailable AND no calendar history exists, fall back to: "No DM or meeting history found. Open with: what's on your mind?"
Context/121s/[Person Name].md — if it exists, read the open loops and recent session log entries first (this is the richest source)Context/Memory/graph.yaml if it exists — find edges where source or target matches this person's node id. Surface:
Context/Meeting Notes/ for recent notes mentioning this personContext/Document Hub/ for shared docs, PRDs, or projects involving themGOALS.md — identify which goals this person is connected to (e.g., key relationships, team members, stakeholders)Context/Memory/ for any stored preferences or notes about working with this personslack_user_id, slack_dm_channel, and slack_canvas_id from the frontmatter to directly fetch recent DMs and read/update the shared canvas — skip searching if these fields are populated. If any field is empty, search Slack to find it and backfill the frontmatter for next time.Tasks/ for active tasks this person is involved in or would care aboutContext/Progress Updates/ for the most recent weekly wrap — what moved, what's blockedOutput a flat list of max 5 bullets. No sections, no headers. Pick the 5 highest-signal items across updates, questions, open loops, and strategic topics. Each bullet must be self-contained — the user will cherry-pick and copy-paste directly into their 1:1 doc. Lead with context, not naked questions.
Maintain a running relationship doc at Context/121s/[Person Name].md. This accumulates context across sessions so future /121 and /today runs start from a richer base.
If the file doesn't exist, create it:
---
title: "121 — [Person Name]"
description: "[Role/team] — ongoing 1:1 relationship doc with talking points, decisions, and open loops"
type: knowledge
topics: ["leadership"]
aliases: []
created_date: YYYY-MM-DD
slack_user_id: "" # Their Slack user ID (e.g. U027T704S5N) — look up via slack_search_users
slack_dm_channel: "" # DM channel ID (e.g. D06RMTHAYU9) — found in Slack search results for DMs
slack_canvas_id: "" # Shared 121 canvas file ID (e.g. F06UNLHE7S6) — search for "121 [name]" canvases
---
# 121 — [Person Name]
## Open Loops
_Unresolved items carried forward across sessions._
## Session Log
On every run, update the file:
[x] with the date### YYYY-MM-DD
**Talking points generated:**
* [bullet from Discuss]
* [bullet from Ask]
* [bullet from Strategic]
**Outcomes:** _Filled in after the meeting by /today or manually._
**New open loops:**
- [ ] [item]
The "Outcomes" field starts empty — it gets filled when the user debriefs after the meeting, or when /today finds evidence in Slack/meeting notes the next day.
If Context/Memory/graph.yaml exists, keep the structured relationship layer in sync:
type: personfile: "Context/121s/[Person Name].md"aliases: pulled from the page frontmatter if presentself -> person edge exists with type: collaborates_withweight by 0.1 (cap 1.0), update last_seen, and append one evidence line like "1:1 YYYY-MM-DD"Present the talking points and ask: "Anything to add or adjust?"
* bullet style (not -) for copy-paste compatibility with Google Docs / Notion.