菜单
Use when onboarding new agents to the team, providing role briefings, conducting project handoffs, and ensuring agents are ready to contribute effectively. Trigger with new agent or new project setup.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/tomevault-io/skills-registry --skill ecos-onboarding命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
| name | ecos-onboarding |
| description | Use when onboarding new agents to the team, providing role briefings, conducting project handoffs, and ensuring agents are ready to contribute effectively. Trigger with new agent or new project setup. |
| user-invocable | false |
| license | Apache-2.0 |
| compatibility | Requires AI Maestro messaging API access, project documentation access, and role definition documents. Requires AI Maestro installed. |
| metadata | {"author":"Emasoft","version":"1.0.0"} |
| context | fork |
| agent | ecos-main |
| workflow-instruction | Step 4 |
| procedure | proc-create-team |
Agent onboarding is the process of integrating new AI agents into a coordinated team. The Chief of Staff ensures that each new agent understands their role, the project context, key procedures, and how to communicate with teammates. Proper onboarding reduces confusion, prevents errors, and accelerates time-to-productivity.
Before using this skill, ensure:
| Onboarding Type | Output |
|---|---|
| New agent | Agent understands role and context |
| New project | Project registered, team assigned |
| Role change | Agent updated with new responsibilities |
Agent onboarding is a structured process to prepare a new agent for effective participation in a multi-agent team. Unlike human onboarding which may take days or weeks, agent onboarding happens in minutes but must be thorough and precise.
Key characteristics:
A systematic list of items to cover during onboarding.
Detailed explanation of the agent's role, responsibilities, and expectations.
Transfer of essential project knowledge, conventions, and current state.
When to use: When a new agent joins the team or when reassigning an agent to a different project.
Steps: Initiate onboarding session, verify agent identity, work through each checklist item, confirm completion, document onboarding.
Related documentation:
When to use: After initial onboarding checklist, when role changes, or when role responsibilities are updated.
Steps: Retrieve role definition, explain responsibilities, clarify reporting structure, set expectations, answer questions, confirm understanding.
Related documentation:
When to use: When onboarding to a specific project, when transferring work between agents, or when bringing backup agents up to speed.
Steps: Provide project overview, share current state, explain conventions, transfer working knowledge, verify comprehension.
Related documentation:
Copy this checklist and track your progress:
Use the agent-messaging skill to send an onboarding initiation message:
new-developer-agentWelcome - Onboarding Session Startinghighrequest, welcoming the agent to the team, introducing yourself as Chief of Staff, and requesting confirmation that the agent is ready to begin onboardingVerify: confirm message delivery and await agent readiness confirmation.
# Role Briefing: Developer
## Your Assigned Role
You are assigned as a **Developer** on the Authentication Module.
## Responsibilities
1. Implement features from the backlog
2. Write unit tests for all new code
3. Update documentation for your changes
4. Participate in code reviews when requested
## Reporting Structure
- **Report to:** orchestrator-master for task assignments
- **Coordinate with:** auth-code-reviewer for reviews
- **Escalate to:** chief-of-staff for blockers
## Expectations
- Acknowledge task assignments promptly
- Provide regular status updates during active work
- Request clarification if requirements are unclear
- Follow project coding conventions
Please confirm you understand these responsibilities.
# Project Handoff: Authentication Module
## Project Overview
We are building a secure authentication module for the main application.
Current sprint: Sprint 5 (2 weeks remaining)
## Current State
- Login endpoint: COMPLETE
- Logout endpoint: IN PROGRESS (60%)
- Password reset: NOT STARTED
- Session management: BLOCKED (waiting on logout)
## Key Files
- src/auth/login.py - Login implementation
- src/auth/logout.py - Logout implementation (your focus)
- tests/auth/ - Test directory
## Conventions
- Use async/await for all I/O operations
- All endpoints return JSON
- Error responses use standard error format
- Tests required for all new functions
## Active Context
The logout endpoint is partially implemented. Current work is at line 145.
Next step: Implement session invalidation.
Step-by-step runbooks for executing each onboarding operation. Use these when performing the actual procedures described above.
Detailed step-by-step runbook for onboarding a new agent to the team, including messaging templates and registry logging.
Detailed step-by-step runbook for delivering a structured role briefing to an agent, including role definition retrieval and understanding confirmation.
Detailed step-by-step runbook for transferring project knowledge to an agent, including state gathering, document composition, and comprehension verification.
Detailed step-by-step runbook for validating handoff documents before sending them to agents, ensuring completeness and accuracy.
Symptoms: No acknowledgment, onboarding cannot begin.
Solution: Verify agent session is active, retry with higher priority, check for messaging issues, escalate to user if persistent.
Symptoms: Agent asks repeated clarification questions, demonstrates confusion.
Solution: Provide more specific examples, break down responsibilities into smaller steps, have agent repeat back understanding, consider if role is appropriate match.
Symptoms: Agent lacks critical context, makes errors based on missing information.
Solution: Review handoff checklist, identify gaps, send supplementary information, verify understanding of each critical point.
Before sending any handoff document to a new agent, validate using this checklist:
### Handoff Validation Checklist
Before sending handoff:
- [ ] All required fields present (from/to/type/UUID/task)
- [ ] UUID is unique (check existing handoffs: `ls $CLAUDE_PROJECT_DIR/thoughts/shared/handoffs/`)
- [ ] Target agent exists and is alive (use the `ai-maestro-agents-management` skill to list agents and verify the target is online)
- [ ] All referenced files exist (`test -f <path> && echo "EXISTS" || echo "MISSING"`)
- [ ] No placeholder [TBD] markers (`grep -r "\[TBD\]" handoff.md`)
- [ ] Document is valid markdown (no broken links, proper formatting)
- [ ] Acceptance criteria clearly defined
- [ ] Current state accurately reflects reality
- [ ] Contact information for questions provided
Validation command:
# Validate the plugin (including handoff documents) using the CPV plugin validator
uv run --with pyyaml python scripts/validate_plugin.py . --verbose
Required fields for all handoffs:
| Field | Description | Example |
|---|---|---|
from | Sending agent name | ecos-chief-of-staff |
to | Target agent name | new-developer-agent |
type | Handoff type | project-handoff, role-briefing, emergency-handoff |
UUID | Unique handoff identifier | HO-20250204-auth-001 |
task | Task or role being handed off | Authentication Module Development |
Version: 1.0 Last Updated: 2025-02-01 Target Audience: Emasoft Chief of Staff Agent Difficulty Level: Intermediate
Converted and distributed by TomeVault — claim your Tome and manage your conversions.
基于 SOC 职业分类
