OpenParable
Skip to content
← Back to skills

Msgraph Teams

Apps & Automation
msgraph-teams

Msgraph Teams is an Agent Skill for apps & automation published in the automateyournetwork/netclaw repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Send notifications and reports to Microsoft Teams channels via Graph API - alert delivery, report posting, incident updates, and diagram sharing. Use when posting health alerts to Teams, sending security notifications, sharing change completion updates, or delivering reports to a Teams channel

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/automateyournetwork/netclaw/blob/HEAD/workspace/skills/msgraph-teams/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/msgraph-teams, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Microsoft Graph — Teams Notifications

How to Call the Tools

The Microsoft Graph MCP server is invoked via npx with Azure AD credentials:

AZURE_TENANT_ID=$AZURE_TENANT_ID AZURE_CLIENT_ID=$AZURE_CLIENT_ID AZURE_CLIENT_SECRET=$AZURE_CLIENT_SECRET \
  python3 $MCP_CALL "npx -y @anthropic-ai/microsoft-graph-mcp" <tool-name> '<arguments-json>'

Available Operations

1. List Teams

AZURE_TENANT_ID=$AZURE_TENANT_ID AZURE_CLIENT_ID=$AZURE_CLIENT_ID AZURE_CLIENT_SECRET=$AZURE_CLIENT_SECRET \
  python3 $MCP_CALL "npx -y @anthropic-ai/microsoft-graph-mcp" graph_list_teams '{}'

2. List Channels in a Team

AZURE_TENANT_ID=$AZURE_TENANT_ID AZURE_CLIENT_ID=$AZURE_CLIENT_ID AZURE_CLIENT_SECRET=$AZURE_CLIENT_SECRET \
  python3 $MCP_CALL "npx -y @anthropic-ai/microsoft-graph-mcp" graph_list_channels '{"teamId":"<team-id>"}'

3. Send a Message to a Channel

AZURE_TENANT_ID=$AZURE_TENANT_ID AZURE_CLIENT_ID=$AZURE_CLIENT_ID AZURE_CLIENT_SECRET=$AZURE_CLIENT_SECRET \
  python3 $MCP_CALL "npx -y @anthropic-ai/microsoft-graph-mcp" graph_send_channel_message '{"teamId":"<team-id>","channelId":"<channel-id>","message":"<html-content>"}'

4. Reply to a Thread

AZURE_TENANT_ID=$AZURE_TENANT_ID AZURE_CLIENT_ID=$AZURE_CLIENT_ID AZURE_CLIENT_SECRET=$AZURE_CLIENT_SECRET \
  python3 $MCP_CALL "npx -y @anthropic-ai/microsoft-graph-mcp" graph_reply_to_message '{"teamId":"<team-id>","channelId":"<channel-id>","messageId":"<msg-id>","message":"<html-content>"}'

5. List Recent Messages

AZURE_TENANT_ID=$AZURE_TENANT_ID AZURE_CLIENT_ID=$AZURE_CLIENT_ID AZURE_CLIENT_SECRET=$AZURE_CLIENT_SECRET \
  python3 $MCP_CALL "npx -y @anthropic-ai/microsoft-graph-mcp" graph_list_channel_messages '{"teamId":"<team-id>","channelId":"<channel-id>"}'

Message Formats

Health Check Alert

<h3>Health Check Report — 2026-02-22</h3>
<table>
  <tr><th>Device</th><th>CPU</th><th>Memory</th><th>Status</th></tr>
  <tr><td>R1</td><td>23%</td><td>45%</td><td>HEALTHY</td></tr>
  <tr><td>SW1</td><td>82%</td><td>67%</td><td>WARNING</td></tr>
</table>
<p><strong>1 WARNING</strong> — SW1 CPU at 82% (threshold: 80%)</p>

Security Alert (Critical)

<h3>CRITICAL — CVE Exposure Detected</h3>
<p><strong>Device:</strong> R1 (IOS-XE 17.9.4a)</p>
<p><strong>CVE-2023-20198</strong> (CVSS 10.0) — Web UI privilege escalation</p>
<p><strong>Exposure:</strong> CONFIRMED — ip http server enabled</p>
<p><strong>Action Required:</strong> Disable ip http server or upgrade immediately</p>

Change Completion

<h3>Change Complete — CHG0012345</h3>
<p><strong>Target:</strong> R1 — Add Loopback99 (99.99.99.99/32)</p>
<p><strong>Status:</strong> SUCCESS — verified, CR closed</p>
<p><strong>GAIT Branch:</strong> config-r1-loopback99-2026-02-22</p>

Topology Diagram Share

<h3>Updated Network Topology</h3>
<p>Campus physical topology regenerated from CDP/LLDP discovery.</p>
<p><strong>Devices:</strong> 4 | <strong>Links:</strong> 6 | <strong>Changes:</strong> +1 new link (R2-SW2)</p>
<p><a href="https://contoso.sharepoint.com/sites/neteng/Topology/campus-physical-2026-02-22.vsdx">Open in Visio</a></p>

When to Use

  • Health alerts — Post health check results with severity-coded summaries
  • Security notifications — Alert on CVE exposure, failed security audits, or posture violations
  • Change updates — Notify the team when changes are completed, failed, or rolled back
  • Incident updates — Post investigation progress and resolution to incident channels
  • Report delivery — Share audit reports, reconciliation results, and compliance summaries
  • Diagram sharing — Post links to updated Visio topology diagrams on SharePoint

Channel Mapping

Map Teams channels to NetClaw notification types:

### Teams Channel Map (configure in TOOLS.md)
- #netclaw-alerts     → P1/P2 critical alerts, CVE exposure
- #netclaw-reports    → Health reports, audit results, reconciliation
- #netclaw-changes    → Change request updates, completion notices
- #network-general    → P3/P4 notifications, topology updates
- #incidents          → Active incident investigation threads

Integration with Other Skills

  • pyats-health-check — Post health check summaries to Teams
  • pyats-security — Alert on CVE exposure and failed security audits
  • netbox-reconcile — Post drift detection results to Teams
  • servicenow-change-workflow — Notify on CR creation, approval, completion
  • ise-incident-response — Post investigation progress to incident channels
  • msgraph-visio — Share Visio diagram links in Teams channels
  • msgraph-files — Reference SharePoint file links in Teams messages
  • gait-session-tracking — Record Teams notifications in audit trail

Rules

  1. Never post credentials or secrets to Teams channels
  2. Always include severity level in alert messages
  3. Thread incident updates — reply to the original message, don't create new top-level posts
  4. Respect channel purpose — alerts go to alert channels, reports go to report channels
  5. Include actionable context — every alert should say what happened and what to do next

GAIT Audit Trail

Record Teams notifications in GAIT:

python3 $MCP_CALL "python3 -u $GAIT_MCP_SCRIPT" gait_record_turn '{"input":{"role":"assistant","content":"Posted fleet health report to Teams #netclaw-reports channel. 4 devices assessed: 3 HEALTHY, 1 WARNING (SW1 CPU 82%). Included table with CPU, memory, and interface status per device.","artifacts":[]}}'