Ollie Bennett

Articles by Ollie Bennett

• OpenAI Agent Builder shut down: move a social data agent to the Agents SDK

  Agent Builder shuts down on Nov 30, 2026. Move a Reddit and YouTube agent to the Agents SDK with MCP, approvals and streaming, from $11.78 a month.

• Claude Code Channels: Push New Competitor Posts Into a Live Session

  What the Claude Code channels docs say, plus a working MCP push server that polls 12 competitor sources hourly and sends only new posts for $8.61 a month.

• Apify n8n alternative: one workflow, two bills

  In n8n, Apify bills per row and EveryInfra bills per call. Full pages cost 29x to 360x less per 1,000 rows. Small daily runs fit in the Apify free credit.

• Gemini Skills Replace Gems: A Weekly Hot Posts Digest

  Gemini Gems go read only on Nov 17. Move to Gemini skills with a weekly hot posts digest: SKILL.md, three data routes and real tests of four hot list APIs.

• Antigravity CLI Comment Monitoring: Setup and Real Data

  Run a 24-hour YouTube and Bilibili comment patrol with Antigravity CLI. Real data: 100 comments covered 5 hours; one day costs $0.023 in API fees.

• Claude Code dynamic workflows for Xiaohongshu comments: API test and script example

  Claude Code dynamic workflows sort Xiaohongshu comments in parallel. Our test: 4 of 8 pages had 10 rows, one post was 96% noise, 500 comments cost $0.222.

• One SKILL.md, Four AI Tools: A Competitor Review Skill

  One competitor review SKILL.md for Claude Code, Codex, Antigravity CLI and Gemini, checked against each official doc. Nine real API calls read 967 reviews for $0.0125.

• Cursor MCP for TikTok, Instagram & Amazon Data: Real Calls

  The EveryInfra plugin is listed on Cursor Directory, and its Cursor Marketplace application is in progress. Three ways to install it, a log of real MCP calls with latency and cost, error meanings and a monthly cost example.

• Unified Data API Getting Started: From Capability Catalog to First Available Result

  Access multi-platform data using platform, action, and params: read the capability catalog, complete minimal requests, handle asynchronous tasks, partial results, and billing, and establish traceable data records.