Cisco Meraki Aam Thousandeyes Setup

Cisco Meraki AAM ThousandEyes Setup

Prerequisites

Workflow Overview

┌───────────┐   ┌───────────────┐   ┌───────────────┐   ┌─────────────────┐
│ Preflight │ → │ Render/review │ → │ Apply/handoff │ → │ Validate evidence │
└───────────┘   └───────────────┘   └───────────────┘   └─────────────────┘

When to Activate

• Link Meraki Dashboard to ThousandEyes, deploy ThousandEyes Enterprise Agents on supported Meraki MX networks, claim or use Meraki AAM free tests, create ThousandEyes tests from Meraki Insight / Active Application Monitoring, monitor an.
• Preview and review the cisco meraki aam thousandeyes setup workflow before any live apply phase.
• Diagnose failed prerequisites, generated assets, configuration, or validation evidence.

Scope

Follow the documented read-only or render-first path whenever it is available. This skill does not imply permission to mutate live systems. Require explicit apply flags, protected credentials, and operator review for state changes.

Examples

Inspect the supported setup modes before selecting one:

bash skills/cisco-meraki-aam-thousandeyes-setup/scripts/setup.sh --help

Expected output: usage, supported modes, and required arguments are displayed without changing the target environment.

Inspect validation modes before running completion checks:

bash skills/cisco-meraki-aam-thousandeyes-setup/scripts/validate.sh --help

Expected output: offline, live, and completion options are displayed when the skill supports them; help exits without mutation.

Troubleshooting

Overview

Use this skill for the Meraki Dashboard side of the Meraki + ThousandEyes Active Application Monitoring (AAM) workflow. It is not the Splunk Meraki TA installer; use cisco-meraki-ta-setup for Splunk_TA_cisco_meraki and cisco-thousandeyes-setup for the Splunk Platform ThousandEyes app.

The Meraki AAM wizard is the supported control plane for provisioning MX-hosted ThousandEyes agents. Public Meraki documentation says Dashboard API agent provisioning is not available, so treat Meraki Dashboard POSTs as private, volatile implementation details. Capture and summarize them for evidence or diagnosis; do not replay them unless the user explicitly asks for private API replay and confirms the exact destination, payload, and side effect.

Safe Workflow

Start with a local intake file and render reviewable artifacts:

cp skills/cisco-meraki-aam-thousandeyes-setup/template.example \
  skills/cisco-meraki-aam-thousandeyes-setup/template.local

bash skills/cisco-meraki-aam-thousandeyes-setup/scripts/setup.sh \
  --render \
  --spec skills/cisco-meraki-aam-thousandeyes-setup/template.local \
  --output-dir meraki-aam-thousandeyes-rendered

If the user exported a browser HAR from the Meraki Dashboard AAM wizard:

bash skills/cisco-meraki-aam-thousandeyes-setup/scripts/setup.sh \
  --render \
  --spec skills/cisco-meraki-aam-thousandeyes-setup/template.local \
  --har ~/Downloads/meraki-aam.har \
  --output-dir meraki-aam-thousandeyes-rendered

The raw HAR can contain cookies, CSRF tokens, org identifiers, and URLs. Never paste raw HAR content into chat. Use the summarizer output instead.

Agent Behavior

• Never ask for Meraki passwords, ThousandEyes passwords, API keys, OAuth tokens, cookies, CSRF values, or HAR files in chat.
• If a user pastes a Meraki API key, ThousandEyes token, cookie, or other credential into chat, do not use it. Tell the user to revoke/rotate it and write the replacement to a chmod-600 local secret file.
• Use browser login flows for account linking. Confirm before clicking Start monitoring, creating tests, unlinking accounts, disconnecting an account, deleting agents, deleting tests, or claiming free tests.
• Prefer the Meraki AAM UI for agent deployment. Use ThousandEyes public APIs only after agents exist and only with a token read from a chmod-600 file.
• Read-only Meraki and ThousandEyes API validation rejects symlinked or hard-linked secret files, opens secrets with no-follow descriptor checks, and pins curl to HTTPS with redirects and user curl configuration disabled.
• Use the Meraki public Dashboard API only for read-only preflight validation such as organizations, networks, and assigned MX devices.
• Do not describe private Meraki Dashboard endpoints as stable or supported. If a captured POST is needed, record method, path, redacted payload shape, response code, and UI step.
• When Chrome access is unavailable, continue with the manual HAR capture workflow in references/har-capture.md.

Workflow

1. Collect non-secret inputs in template.local: Meraki organization, region, target application, desired test profile, network names/IDs, MX serials, and any non-secret ThousandEyes account group ID.
2. Read reference.md for current product boundaries, prerequisites, and source-backed limitations. Read references/har-capture.md before inspecting browser requests or exported HAR files.
3. Preflight eligibility:
   • Meraki account has at least two full organization admins.
   • ThousandEyes user is an Account Admin and can use local auth for linking.
   • Meraki and ThousandEyes account regions are compatible.
   • Target networks use supported MX models, MX 18.104 or later, NAT mode, and working HTTPS reachability to ThousandEyes plus registry.meraki-applications.com.
   • Licensing supports the intended action: SD-WAN+ for agent installation and test deployment; Advanced Security for agent installation only unless ThousandEyes units are purchased separately.
4. Render the plan with scripts/setup.sh --render and review:
   • meraki-aam-plan.md for the UI deployment sequence.
   • browser-capture-checklist.md for request-capture steps.
   • thousandeyes-validation.md for post-deployment validation.
   • handoff-thousandeyes-api-tests.yaml if direct ThousandEyes API test creation is preferred after MX-hosted agents exist.
5. Operate the Meraki Dashboard UI:
   • Go to Insight > Active Application Monitoring.
   • Start the ThousandEyes onboarding flow (Try it, Get started, or equivalent current UI).
   • Link the ThousandEyes account and select the account group if prompted.
   • Select a verified application template or custom application target.
   • Select eligible Meraki networks where the MX-hosted agents will run.
   • Stop before the final side-effecting button and confirm the action with the user before submitting.
6. Validate after deployment. Meraki docs say reports can take up to 15 minutes to appear after agent configuration.

Read-only Meraki API preflight with a rotated key file:

bash skills/shared/scripts/write_secret_file.sh /tmp/meraki_api_key

bash skills/cisco-meraki-aam-thousandeyes-setup/scripts/validate.sh \
  --meraki-api-key-file /tmp/meraki_api_key \
  --meraki-org-id "123456" \
  --network-filter "Branch 001" \
  --output-dir meraki-aam-thousandeyes-rendered/live-validation

ThousandEyes validation after the Meraki wizard starts monitoring:

bash skills/cisco-meraki-aam-thousandeyes-setup/scripts/validate.sh \
  --te-token-file /path/to/te_token \
  --account-group-id "1234" \
  --agent-filter "branch-or-mx-name" \
  --test-filter "application-name" \
  --output-dir meraki-aam-thousandeyes-rendered/live-validation

Creating Tests

For the Meraki AAM experience, create tests through the Meraki wizard when the requested test is represented by the UI template or custom target flow. Capture POSTs only to understand what the UI did.

For public ThousandEyes API test lifecycle after the Meraki MX agents exist, hand off handoff-thousandeyes-api-tests.yaml to splunk-observability-thousandeyes-integration or use the official ThousandEyes Tests API directly. The rendered handoff intentionally leaves agent IDs as TODO_REPLACE_WITH_MERAKI_AGENT_IDS until validation confirms the correct MX-hosted Enterprise Agents.

Resources

• template.example - local intake worksheet for non-secret AAM inputs.
• scripts/setup.sh - renders reviewable AAM runbooks and optional HAR summaries.
• scripts/summarize_har.py - redacts and summarizes exported HAR files.
• scripts/validate.sh - validates Meraki organizations/networks/MX devices and ThousandEyes agents/tests through public API calls using secret files.
• reference.md - source-backed Meraki/ThousandEyes requirements and API boundaries.
• references/har-capture.md - browser capture and redaction protocol.

Hand-offs

• Splunk Platform ThousandEyes ingestion and dashboards: cisco-thousandeyes-setup.
• Splunk Platform Meraki telemetry ingestion: cisco-meraki-ta-setup.
• ThousandEyes MCP registration: cisco-thousandeyes-mcp-setup.
• TE API-backed tests, alert rules, streams, and Splunk Observability handoffs: splunk-observability-thousandeyes-integration.