FAQ
Frequently asked questions about Flezi AgentBox platform usage, account management, and troubleshooting
Frequently Asked Questions
Find answers to common questions about the Flezi AgentBox platform. Can't find what you're looking for? Use the feedback widget to submit your question.
Account & Authentication
How do I create an account?
Navigate to the sign-up page and register with your email address. You'll receive a verification email to activate your account.
I forgot my password. How do I reset it?
Click Forgot Password on the login page. Enter your email and follow the instructions in the reset email.
Can I use Flezi AgentBox without an account?
Yes — you can browse the marketplace and view agent details as a guest. However, creating agents, executing agents, and interacting with the community requires an account.
How do I delete my account?
Contact support through the feedback widget. Account deletion removes all your data, agents, and execution history permanently.
API Keys
How do I get an API key?
Navigate to Settings > API Keys and click Generate New Key. Copy the key immediately — it won't be shown again.
Are my API keys secure?
Yes. API keys are encrypted using AES-256-GCM server-side encryption before storage. They are never stored in plain text and cannot be viewed after creation.
Can I have multiple API keys?
Yes. You can create multiple keys for different applications or environments. Each key can be individually revoked.
What happens if my API key is compromised?
Immediately revoke the key in Settings > API Keys and generate a new one. Review your execution logs for unauthorized usage.
Agent Creation
How long does it take to build an agent?
A simple agent can be built in under 5 minutes using a template. Complex agents with custom pipelines may take longer depending on your requirements.
Can I use any LLM provider?
Flezi AgentBox supports OpenAI, Anthropic, and Google LLM providers. You configure your preferred provider in Settings and can override per-agent.
What is the BMAD Expansion Pack?
The BMAD Expansion Pack is an advanced feature that generates complete BMAD workflow packs from your agent configuration. It's designed for teams using the BMAD methodology.
Can I fork another creator's agent?
Yes, if the agent allows forking. Forked agents include attribution to the original creator and maintain a link to the source.
Execution
How are agents executed?
Agents run in isolated Docker containers on the Flezi AgentBox orchestrator. Each execution gets its own container with CPU, memory, and time limits.
What are the execution limits?
Default limits: 5 minutes timeout, 512MB memory, 1 CPU core. These may vary based on your plan and the agent's configuration.
Can I schedule agent executions?
Yes. Use webhook triggers or cron scheduling to automate agent executions. Configure these in the agent's settings.
Why did my execution fail?
Common causes include:
- Timeout — The agent took too long to respond
- API Key Error — Your LLM provider key is invalid or expired
- Rate Limit — Too many requests to the LLM provider
- Pipeline Error — A node in the pipeline produced invalid output
Check the execution logs in your dashboard for detailed error information.
Marketplace
How do I publish an agent?
Complete the Builder Lab workflow (Discovery → Logic → Synthesis) and click Publish. Your agent will appear in the marketplace immediately.
Can I set pricing for my agents?
Yes. In the agent settings, configure per-execution pricing. Revenue is shared between you and the platform.
How does the Impact Score work?
The Impact Score combines stars (30%), completion rate (25%), repeat usage (25%), and review average (20%). Higher scores improve marketplace visibility.
What is Flezi AgentBox Verified certification?
Agents that pass a quality review and have an automated test suite can earn the Flezi AgentBox Verified badge, which signals quality to users.
Pipelines
What's the difference between an agent pipeline and an agent's internal pipeline?
An agent's internal pipeline (built in the Logic phase) defines processing steps within a single agent. An agent pipeline chains multiple published agents together.
Can pipelines run agents in parallel?
Yes. The pipeline builder supports both sequential and parallel execution patterns. Independent stages can run simultaneously.
IDE Integration
Which IDEs are supported?
VS Code, Cursor, and Claude Code are supported via the Model Context Protocol (MCP) integration.
Do I need the platform running to use IDE integration?
Yes. The MCP server runs as part of the Flezi AgentBox orchestrator. Your IDE connects to it over HTTP.
Billing & Pricing
Is Flezi AgentBox free to use?
Core features are free, including account creation, agent browsing, and limited executions. Premium features and high-volume usage may require payment.
How do creator payouts work?
Creators earn revenue from paid agent executions. View earnings on the Creator Payout Dashboard in Settings.
Troubleshooting
The platform is slow or unresponsive
- Clear your browser cache and refresh
- Check your network connection
- Try a different browser
- If the issue persists, submit feedback via the widget
My agent won't publish
Ensure all three Builder Lab phases are complete and the pipeline passes validation. Check for any error messages in the Synthesis phase.
I can't see my agent in the marketplace
Verify the agent is published (not draft) and has public visibility. New agents may take a few seconds to appear in search results.