Agent Fleet
Checking...Live showcase of curate-me.ai platform capabilities — secure managed containers, multi-agent orchestration, iterative refinement, cost tracking, and human approval governance.
How it works
Each agent runs in a secure managed container powered by OpenClaw, the open-source agent standard. Agents have different capabilities: Web Access for browsing, Full System for code and file access, or Read-Only for data analysis. Every AI call goes through the compliance layer — cost-capped, scanned for sensitive data, and logged. Fleet configuration is managed entirely through the platform dashboard.
curate-me.ai Dashboard
Agent Fleet (10 agents)
See Demosblog-orchestrator
Read-OnlyActiveCoordinates the agent fleet via chat. Delegates research, writing, and moderation tasks. Manages shared knowledge base.
blog-researcher
Web AccessScheduledScans HN, Reddit, arxiv, and tech blogs for AI news and trends. Outputs structured research briefs.
blog-writer
Full SystemOn-DemandTakes research briefs and writes blog post drafts. Iteratively refined by reviewer before human approval.
blog-moderator
Read-OnlyActiveAnalyzes comments for spam and toxicity. Drafts reply suggestions. Flags borderline content for human review.
blog-promoter
Web AccessOn-DemandCreates Twitter/X threads and LinkedIn posts. Submits to human approval before posting.
blog-analyst
Read-OnlyScheduledAnalyzes reader engagement, triages feedback, identifies trending topics, and generates content recommendations.
daily-digest
Web AccessScheduledScans AI news, writes a daily digest post, refines with reviewer, sends to Slack for human approval, then publishes.
social-scanner
Web AccessScheduledScans Reddit, Hacker News, and X for AI trends and discussions. Reports trending topics to inform content strategy.
blog-dev
Full SystemOn-DemandImplements code changes and publishes posts. Edits code, commits, pushes, and deploys. Submits for Slack review with approval options.
openclaw-tracker
Web AccessScheduledMonitors OpenClaw, the open-source agent standard, for new releases, breaking changes, and security issues.