Examples

API workflows teams actually build

Use these patterns as starting points for internal dashboards, weekly reporting, content monitoring, citation research, and assistant workflows.

Weekly executive scorecard

Pull headline visibility, citation rate, sentiment, and share of voice for the last 30 days.

curl "https://api.llmpulse.ai/api/v1/metrics/summary?project_id=123&metrics=mentions,mention_rate,citation_rate,weighted_visibility,net_sentiment&range=30" \
  -H "Authorization: Bearer llmpulse_your_api_key_here"

Prompt-level opportunity list

Sort prompts by responses or visibility to find topics where your brand is absent or weak.

curl "https://api.llmpulse.ai/api/v1/metrics/prompt_summary?project_id=123&range=30&sort=mention_rate&sort_dir=asc&per_page=50" \
  -H "Authorization: Bearer llmpulse_your_api_key_here"

Citation source audit

Find the domains and pages that AI answers cite most often for your tracked topics.

curl "https://api.llmpulse.ai/api/v1/citation_intelligence/groups?project_id=123&group_by=domain&range=30" \
  -H "Authorization: Bearer llmpulse_your_api_key_here"

Search inside AI answers

Use query mode to count responses that mention a brand, product, person, or phrase.

curl "https://api.llmpulse.ai/api/v1/answers?project_id=123&query=pricing&range=30" \
  -H "Authorization: Bearer llmpulse_your_api_key_here"

Integration Patterns

Most customer builds fit one of these four shapes.

BI Dashboard

Pull /metrics/timeseries, /metrics/summary, and /metrics/sov into a warehouse or BI tool.

Content Queue

Combine prompt summaries, low-visibility topics, top cited sources, and recommendations into a weekly content backlog.

Alerting

Compare current visibility or sentiment against a saved baseline, then notify the team when important prompts move.

AI Assistant

Use MCP so teams can ask natural-language questions about projects, competitors, citations, recommendations, and prompt results.

Webhook Flow

Webhooks send signed project events to your HTTPS endpoint. Use them for automations where polling is wasteful.

curl -X POST "https://api.llmpulse.ai/api/v1/webhooks" \
  -H "Authorization: Bearer llmpulse_your_api_key_here" \
  -H "Content-Type: application/json" \
  -d '{
    "project_id": 123,
    "event": "prompt_execution.completed",
    "url": "https://example.com/webhooks/llmpulse"
  }'

Metrics Cheat Sheet

Useful fields when building your first dashboard.

Metric Meaning Typical use
mentions Number of responses where the brand was mentioned. Volume and trend monitoring.
mention_rate Mentions divided by responses, expressed from 0 to 100. Primary visibility rate.
weighted_visibility Position-weighted visibility, also accepted as ai_visibility_score. Prominence, not only presence.
citations Count of cited URLs for your brand or topic context. Source strategy and content proof.
citation_rate Citations divided by responses, expressed from 0 to 100. How often answers include source support.
net_sentiment Positive sentiment count minus negative sentiment count. Reputation trend and risk tracking.

Next Steps

Move from example calls to a working integration.

1. Choose a project

Call /dimensions/projects and store the project ID you want to report on.

2. Pick a reporting range

Use range=30 for first dashboards, then move to exact from and to values when needed.

3. Save request IDs

Log request_id from API responses. It helps support trace a request if a dashboard number looks off.