# Onboarding Plan > Design structured onboarding plans — for new hires, new customers, or new users — with milestone-based progression, success criteria, and feedback loops. - **Type:** Skill - **Install:** `agentstack add skill-the-ai-directory-company-agents-and-skills-onboarding-plan` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [The-AI-Directory-Company](https://agentstack.voostack.com/s/the-ai-directory-company) - **Installs:** 0 - **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml) - **Latest version:** 0.1.0 - **License:** MIT - **Upstream author:** [The-AI-Directory-Company](https://github.com/The-AI-Directory-Company) - **Source:** https://github.com/The-AI-Directory-Company/agents-and-skills/tree/main/skills/onboarding-plan - **Website:** https://ai-directory.company ## Install ```sh agentstack add skill-the-ai-directory-company-agents-and-skills-onboarding-plan ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # Onboarding Plan ## Before you start Gather the following from the user: 1. **Who is being onboarded?** (New hire, new customer, or new user — and their role/persona) 2. **What does "fully ramped" look like?** (3-5 concrete outcomes that indicate onboarding is complete) 3. **What is the expected ramp timeline?** (30 days, 60 days, 90 days) 4. **Who is responsible for the onboarding?** (Manager, buddy, CSM, or self-guided) 5. **What existing resources are available?** (Documentation, training videos, sandbox environments, mentors) If the user says "we just need a checklist," push back: "A checklist gets tasks done but doesn't ensure understanding. What should this person be able to do independently by the end of onboarding?" ## Onboarding plan template ### 1. Onboarding Overview Define the onboarding scope in a brief summary. ``` Onboardee type: New hire — Senior Frontend Engineer Team: Consumer Products Timeline: 90 days (30-60-90 structure) Onboarding owner: Engineering Manager + assigned buddy Success measure: Ships first production feature independently by Day 75 ``` ### 2. Pre-Start Preparation Complete before the onboardee's first day: - **Access provisioning**: List every system, tool, and repo they need access to. Include account names and who provisions each one. - **Equipment setup**: Laptop, monitors, peripherals — ordered and configured. - **Welcome packet**: Team org chart, key contacts, calendar invites for recurring meetings, links to onboarding docs. - **Buddy assignment**: Assign a peer (not the manager) who is available for daily questions during the first two weeks. - **First-week calendar**: Pre-schedule all Day 1-5 meetings and activities. An empty calendar on Day 1 signals disorganization. ### 3. Milestone-Based Progression Structure the plan into milestones, not a flat task list. Each milestone has a clear success criterion. **Milestone 1: Orientation (Days 1-7)** | Day | Activity | Owner | Success Criterion | |-----|----------|-------|--------------------| | 1 | Welcome meeting with manager — role expectations, 90-day goals | Manager | Goals documented and shared | | 1 | Dev environment setup with buddy | Buddy | Can build and run the app locally | | 2 | Architecture walkthrough — system overview, key services | Tech lead | Can draw the high-level architecture from memory | | 3 | Codebase tour — repo structure, CI/CD pipeline, deployment process | Buddy | Successfully deploys a test change to staging | | 4-5 | Read team docs — ADRs, runbooks, on-call playbook | Self | Completes reading, notes questions for buddy | **Milestone 2: Guided Contribution (Days 8-30)** | Week | Activity | Owner | Success Criterion | |------|----------|-------|--------------------| | 2 | Pick up first "good first issue" ticket | Manager | PR submitted with tests | | 2-3 | Pair programming sessions (2-3 per week) | Buddy | Onboardee drives, buddy observes | | 3 | Attend sprint planning and retro — observe, don't commit yet | Self | Understands team process | | 4 | Ship first bug fix or small feature to production | Self | Code reviewed and merged | | 4 | **30-day check-in with manager** | Manager | Written feedback exchanged both ways | **Milestone 3: Independent Contribution (Days 31-60)** | Week | Activity | Owner | Success Criterion | |------|----------|-------|--------------------| | 5-6 | Own a medium-sized feature end-to-end | Manager | Writes own technical approach, gets feedback | | 6-7 | Participate in code review for others | Self | Provides substantive review comments | | 7-8 | Shadow on-call rotation (if applicable) | On-call lead | Can handle a low-severity alert independently | | 8 | **60-day check-in with manager** | Manager | Aligned on remaining ramp goals | **Milestone 4: Full Ramp (Days 61-90)** | Week | Activity | Owner | Success Criterion | |------|----------|-------|--------------------| | 9-10 | Ship a production feature independently | Self | Scoped, built, tested, deployed without hand-holding | | 10-11 | Present a technical topic to the team | Self | Demonstrates domain understanding | | 12 | Join on-call rotation (if applicable) | Manager | Added to rotation schedule | | 12 | **90-day review** | Manager | Formal assessment against 90-day goals | ### 4. Feedback Loops Build feedback into the plan — don't wait for the end. - **Daily (Week 1)**: 15-minute buddy check-in. One question: "What's blocking you?" - **Weekly (Weeks 2-4)**: 30-minute 1:1 with manager. Review progress against milestones. - **Biweekly (Weeks 5-12)**: Standard 1:1 cadence. Shift from onboarding topics to regular work. - **Formal checkpoints**: Written feedback at 30, 60, and 90 days. Both sides share what's working and what isn't. At each checkpoint, ask the onboardee: "What's one thing about the onboarding that should change for the next person?" ## Quality checklist Before delivering the plan, verify: - [ ] Every milestone has measurable success criteria, not just activities - [ ] Pre-start preparation is complete — no "figure it out on Day 1" gaps - [ ] A buddy is assigned and their responsibilities are documented - [ ] Feedback checkpoints are scheduled at 30, 60, and 90 days - [ ] The plan distinguishes between "do this task" and "be able to do this independently" - [ ] Access provisioning lists specific systems, not "request access as needed" - [ ] The 90-day completion criteria define what "fully ramped" means concretely ## Common mistakes to avoid - **Task lists without success criteria.** "Read the architecture docs" is a task. "Can draw the system architecture from memory and explain data flow" is a success criterion. Every activity needs a way to verify understanding. - **No buddy assignment.** Managers are too busy for daily questions. A buddy at the same level provides low-friction support. Assign one explicitly and protect their time. - **Information firehose on Day 1.** Spreading 8 hours of presentations across the first day guarantees nothing is retained. Limit Day 1 to setup, one key meeting, and one hands-on activity. - **No feedback until 90 days.** If something is off at Week 2, waiting until Day 90 wastes everyone's time. Build in explicit checkpoints at 30 and 60 days with written feedback. - **Treating onboarding as one-size-fits-all.** A senior hire needs less hand-holding on tools but more context on team dynamics and decision history. Adjust the plan to the person's level and background. ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [The-AI-Directory-Company](https://github.com/The-AI-Directory-Company) - **Source:** [The-AI-Directory-Company/agents-and-skills](https://github.com/The-AI-Directory-Company/agents-and-skills) - **License:** MIT - **Homepage:** https://ai-directory.company Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-the-ai-directory-company-agents-and-skills-onboarding-plan - Seller: https://agentstack.voostack.com/s/the-ai-directory-company - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.