octolens/skill

Overview

This skill connects to the Octolens API to query and analyze brand mentions across social platforms. It fetches mentions, applies powerful filters, and returns structured results including sentiment, author metrics, tags, and engagement data. Use it to monitor brand visibility, track keyword performance, and surface actionable social insights.

How this skill works

You provide an API key and filter parameters, and the skill calls the /mentions, /keywords, or /views endpoints on the Octolens API. It supports simple AND-based filters and an advanced groups/operator model for AND/OR logic, plus exclusions, date ranges, follower thresholds, and pagination via cursor. Responses include metadata (sentiment, source, authorFollowers, tags) so the agent can summarize trends or drill into individual posts.

When to use it

• Fetch recent brand mentions across Twitter, Reddit, GitHub, LinkedIn, YouTube and more
• Filter mentions by sentiment, language, follower counts, or engagement status
• Build complex queries using AND/OR groups and exclusions to isolate relevant signals
• List configured keywords or saved views before running targeted queries
• Paginate and aggregate results for dashboards, reports, or alerts

Best practices

• Always ask the user for their Octolens API key before making requests and store it for the session
• Start with a simple filter set, then escalate to advanced groups when you need complex logic
• Use saved views to reuse common filter configurations and reduce query errors
• Respect rate limits (500 requests/hour) and check X-RateLimit-* headers before bulk requests
• Request keyword IDs from /keywords before filtering by keyword and use cursor pagination for large result sets

Example use cases

• Show positive Twitter mentions in the last 7 days with authorFollowers ≥ 1000
• List Reddit and GitHub mentions excluding posts tagged as spam or irrelevant
• Run a composite query: (Twitter OR LinkedIn) AND positive sentiment in the past week
• Fetch all bookmarked mentions for a saved view called 'High Priority' and summarize themes
• Generate a weekly engagement report: top authors, sentiment trend, and platform distribution