Grafana
全球最大的 Claude Code 技能聚合库 · 收录 3900+ 来自 12+ 来源的技能,提供在线搜索与趋势分析看板 / The world's largest Claude Code skill aggregation hub — 3900+ skills from 12+ sources with online search and trend dashboard
npx -y skills add bg-szy/TOP-SKILLS --skill grafanaAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
- 4 stars4 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Comprehensive skill for interacting with Grafana's HTTP API to manage dashboards, data sources, folders, alerting, annotations, users, teams, and organizations. Use when Claude needs to (1) Create, read, update, or delete Grafana dashboards, (2) Manage data sources and connections, (3) Configure alerting rules, contact points, and notification policies, (4) Work with folders and permissions, (5) Manage users, teams, and service accounts, (6) Create or query annotations, (7) Execute queries against data sources, or any other Grafana automation task via API.
SKILL.md
7.7 KB, as published. Nobody here has run it
grafana-skill
Programmatically manage Grafana resources using TypeScript tools and HTTP API workflows.
Workflow Routing
| Workflow | Trigger | File |
|---|---|---|
| DashboardCrud | "create dashboard", "update dashboard", "delete dashboard", "list dashboards", "export dashboard" | Tools/DashboardCrud.ts |
| GrafanaClient | "grafana API", "grafana client", "TypeScript grafana" | Tools/GrafanaClient.ts |
| ApiReference | "grafana API reference", "grafana endpoints" | References/ |
Tools
DashboardCrud CLI
# Set environment variables
export GRAFANA_URL="https://grafana.example.com"
export GRAFANA_TOKEN="your-service-account-token"
# List dashboards
bun run Tools/DashboardCrud.ts list
bun run Tools/DashboardCrud.ts list --query production --tag monitoring
# Get dashboard by UID
bun run Tools/DashboardCrud.ts get abc123
# Export dashboard to JSON
bun run Tools/DashboardCrud.ts export abc123 --output dashboard.json
# Create dashboard from JSON file
bun run Tools/DashboardCrud.ts create --file dashboard.json --folder my-folder
# Update dashboard
bun run Tools/DashboardCrud.ts update abc123 --file updated.json --message "Updated panels"
# Clone dashboard
bun run Tools/DashboardCrud.ts clone abc123 --title "Production Copy" --folder prod-folder
# View version history
bun run Tools/DashboardCrud.ts versions abc123
# Restore to previous version
bun run Tools/DashboardCrud.ts restore abc123 --version 5
# Delete dashboard
bun run Tools/DashboardCrud.ts delete abc123
GrafanaClient TypeScript Library
import { GrafanaClient, createGrafanaClient } from './Tools/GrafanaClient';
// Initialize from environment variables
const client = createGrafanaClient();
// Or with explicit config
const client = new GrafanaClient({
baseUrl: 'https://grafana.example.com',
token: 'your-service-account-token',
orgId: 1, // optional
});
// Dashboard operations
const dashboards = await client.searchDashboards({ query: 'production', tag: 'monitoring' });
const dashboard = await client.getDashboardByUid('abc123');
const saved = await client.saveDashboard({ dashboard: myDashboard, folderUid: 'folder-uid' });
await client.deleteDashboard('abc123');
// Version management
const versions = await client.getDashboardVersions('abc123');
await client.restoreDashboardVersion('abc123', 5);
// Folders, Data sources, Alerting, Annotations also available
Authentication
# Service Account Token (Recommended)
export GRAFANA_TOKEN="glsa_xxxxxxxxxxxxxxxxxxxx"
# Multi-Organization Header
curl -H "Authorization: Bearer $GRAFANA_TOKEN" \
-H "X-Grafana-Org-Id: 2" \
https://grafana.example.com/api/org
Quick API Reference
| Resource | List | Get | Create | Update | Delete |
|---|---|---|---|---|---|
| Dashboards | GET /api/search | GET /api/dashboards/uid/:uid | POST /api/dashboards/db | POST /api/dashboards/db | DELETE /api/dashboards/uid/:uid |
| Folders | GET /api/folders | GET /api/folders/:uid | POST /api/folders | PUT /api/folders/:uid | DELETE /api/folders/:uid |
| Data Sources | GET /api/datasources | GET /api/datasources/uid/:uid | POST /api/datasources | PUT /api/datasources/uid/:uid | DELETE /api/datasources/uid/:uid |
| Alert Rules | GET /api/v1/provisioning/alert-rules | GET /api/v1/provisioning/alert-rules/:uid | POST /api/v1/provisioning/alert-rules | PUT /api/v1/provisioning/alert-rules/:uid | DELETE /api/v1/provisioning/alert-rules/:uid |
Reference Documentation
- Dashboards: Complete dashboard CRUD, versions, permissions
- DataSources: Data source management, queries, health checks
- Alerting: Alert rules, contact points, notification policies
- Folders: Folder management and permissions
- Annotations: Create, query, update annotations
- UsersTeams: User management, team operations
- CommonPatterns: Error handling, pagination, utilities
Examples
Example 1: List and export dashboards
User: "List all production dashboards and export them"
→ bun run Tools/DashboardCrud.ts list --tag production
→ For each: bun run Tools/DashboardCrud.ts export <uid>
→ Returns list of exported JSON files
Example 2: Create dashboard from JSON
User: "Create a new dashboard from this JSON file"
→ bun run Tools/DashboardCrud.ts create --file dashboard.json --folder monitoring
→ Returns new dashboard UID and URL
Example 3: Clone dashboard to another folder
User: "Clone the CPU dashboard to the production folder"
→ bun run Tools/DashboardCrud.ts clone cpu-uid --title "CPU Prod" --folder prod-folder
→ Returns cloned dashboard details
Example 4: Restore dashboard version
User: "Restore dashboard abc123 to version 5"
→ bun run Tools/DashboardCrud.ts versions abc123
→ bun run Tools/DashboardCrud.ts restore abc123 --version 5
→ Dashboard restored, new version created
Example 5: Programmatic bulk update
User: "Write TypeScript to bulk update dashboard tags"
→ Uses GrafanaClient library:
const client = createGrafanaClient();
const dashboards = await client.searchDashboards({ tag: 'old-tag' });
for (const dash of dashboards) {
const full = await client.getDashboardByUid(dash.uid);
full.dashboard.tags = full.dashboard.tags.filter(t => t !== 'old-tag');
full.dashboard.tags.push('new-tag');
await client.saveDashboard({ dashboard: full.dashboard, message: 'Updated tags' });
}
Error Handling
| Code | Description |
|---|---|
| 200 | Success |
| 400 | Bad request (invalid JSON, missing required fields) |
| 401 | Unauthorized (invalid/missing token) |
| 403 | Forbidden (insufficient permissions) |
| 404 | Not found |
| 409 | Conflict (resource already exists) |
| 412 | Precondition failed (version mismatch) |
Tips
- Use UIDs over IDs: UIDs are portable across Grafana instances
- Include version for updates: Prevents overwriting concurrent changes
- Use
overwrite: truecarefully: Only when you want to force-update - Service accounts over API keys: API keys are deprecated in newer Grafana versions
Gotchas
- Datasource permission: viewers can see the panel but get "query forbidden" if the datasource has explicit per-team permissions — confusing for non-admin users.
- Variable interpolation:
$__intervaland$__nameare interpreted differently in legacy vs newer query backends; same dashboard JSON behaves differently. - Provisioning datasources via YAML under
/etc/grafana/provisioning: edits via UI are silently reverted on file re-read; pick one source of truth. - API tokens are per-org — multi-org Grafana setups need separate tokens; a global "admin" token doesn't span orgs.
- Annotation queries with missing time-field return zero annotations without error — debug by checking the raw query response, not the panel.
- Alert rule state lives in the Grafana DB, not in Prometheus — rolling Grafana back loses alert state history.