Skip to main content

Track Brand Mentions on X: Your AI Agent Builds the Weekly Report

· 5 min read
Felo Search Tips Buddy
Felo Search Tips Buddy
Committed to answers at your fingertips

Step-by-step tutorial: connect your agent to Felo OpenAPI, then automate brand monitoring on X with sentiment analysis and weekly presentation reports.

hero

Every Monday, you need to know what people said about your brand on X last week. Instead of manually searching, copying, and compiling, your agent does it automatically.

Here's exactly how.

3 Minutes to Connect Felo OpenAPI​

Step 1: Go to openapi.felo.ai and create a free API key.

Step 2: Copy and paste this into your Codex, Claude Code, or OpenClaw:

Connect to Felo OpenAPI:
- Models: Claude Opus 4.8, Claude Sonnet 4.6, Claude Haiku 4.5, GPT-5.6 Sol, GPT-5.6 Terra, Grok 4.5
- Search: POST /v2/chat (AI Search), POST /v2/web/extract (Web Fetch)
- Social: POST /v2/x/tweet/search (X Search), POST /v2/x/user/tweets (User Tweets), GET /v2/youtube/subtitling
- Output: POST /v2/ppts (PPT), POST /v2/mindmap, POST /v2/landing_page, Image Generation
- Memory: LiveDocs (semantic retrieval, file upload, URL resources, tasks, records)
Base URL: https://openapi.felo.ai
API Key: [YOUR_API_KEY]

When I ask for brand monitoring:
1. Use X Search to find brand mentions
2. Analyze sentiment (positive, negative, neutral)
3. Compare with previous reports in LiveDocs
4. Generate a summary report and presentation

Step 3: Connection complete. Your agent can now monitor social signals, analyze sentiment, and produce reports.

The User Story​

You're the brand manager for an AI tool. Every Monday morning, your boss asks: "What did people say about us on X last week?"

You need:

  • Total mention count
  • Sentiment breakdown (positive/negative/neutral)
  • Key discussions and KOL mentions
  • Trend comparison with the previous week
  • A presentation-ready summary

You Only Need to Say This​

Search X for mentions of "Felo AI" from the past week. Analyze sentiment, identify key discussions and KOL mentions, compare with last week's data in LiveDocs, and generate a weekly report presentation.

What Codex Does Automatically​

Step 1: Search for Brand Mentions​

POST /v2/x/tweet/search
{
  "query": "\"Felo AI\" OR \"felo.ai\"",
  "sort": "relevance",
  "time_range": "7d"
}

Codex pulls all tweets mentioning the brand, including retweets and replies.

Step 2: Search for Competitor Mentions (Context)​

POST /v2/x/tweet/search
{
  "query": "competitor brand name",
  "sort": "relevance",
  "time_range": "7d"
}

Codex also pulls competitor mentions for context — how does your brand's share of voice compare?

Step 3: Analyze Sentiment​

Codex reads through the tweets and categorizes them:

  • Positive: Praise, recommendations, positive user experiences
  • Negative: Complaints, bugs, feature requests, criticism
  • Neutral: News, factual mentions, neutral discussions

It also identifies:

  • Top KOL accounts that mentioned the brand
  • Most-engaged tweets
  • Emerging discussion themes

Step 4: Compare with Last Week​

Codex retrieves last week's report from LiveDocs:

POST /resources/retrieve
{
  "query": "brand monitoring report",
  "filters": { "date_range": "previous_week" }
}

It compares:

  • Mention volume (up/down percentage)
  • Sentiment shift (more positive? more negative?)
  • New discussion themes
  • New KOL mentions

Step 5: Generate the Weekly Report​

POST /v2/ppts
{
  "topic": "Brand Monitoring Weekly Report",
  "content": {
    "total_mentions": [count],
    "sentiment_breakdown": { "positive": X%, "negative": X%, "neutral": X% },
    "top_discussions": [key themes],
    "kol_mentions": [notable accounts],
    "week_over_week": [trend comparison],
    "action_items": [recommended responses]
  },
  "theme": "executive-summary"
}

Returns: A clean, executive-ready weekly report presentation.

Step 6: Save to LiveDocs​

Codex saves this week's report to LiveDocs with:

  • Raw mention data with tweet URLs
  • Sentiment analysis results
  • KOL profiles and engagement metrics
  • The generated presentation
  • Tags for date, brand, and report type

Next week's report automatically builds on this data.

The Result​

Every Monday, you have:

  • Dashboard: Total mentions, sentiment breakdown, trend direction
  • Key Discussions: What people are actually saying
  • KOL Watch: Who's talking about your brand and their reach
  • Week-over-Week: What changed since last week
  • Action Items: Recommended responses or follow-ups
  • Presentation: Ready to share in the team meeting

The Complete Prompt Template (Copy & Reuse)​

Search X for mentions of "[BRAND]" from the past week.

Analyze:
1. Total mention count and sentiment breakdown
2. Top 5 discussions and themes
3. Notable KOL mentions with engagement metrics
4. Comparison with last week's data (stored in LiveDocs)
5. Recommended action items

Generate an executive summary presentation and save all data to LiveDocs.

Replace [BRAND] for any brand monitoring need. You can also add competitor names for share-of-voice analysis.

Advanced: Multi-Brand Dashboard​

Monitor multiple brands at once:

Search X for mentions of these brands from the past week: "Brand A", "Brand B", "Brand C". Produce a comparative dashboard showing share of voice, sentiment by brand, and cross-brand discussion themes. Generate a presentation.

Advanced: Real-Time Alerting​

Set up a daily check:

Search X for mentions of "[BRAND]" from the last 24 hours. If negative sentiment exceeds 30%, flag it as an alert and summarize the key complaints. Otherwise, just log the data to LiveDocs.

Getting Started​

  1. Create your free API key
  2. Paste the initialization prompt into your agent
  3. Try the brand monitoring prompt above

Stop manually compiling social reports. Let your agent monitor, analyze, and present.

Get Your Free API Key →