Quick Start — 5 Minutes
In this guide you'll build a weekly competitor content summariser: every Monday morning, an AI agent reads competitor blog posts, summarises each one into 3 bullet points, and saves the results as a downloadable artifact.
No LLM API key? You can use a local agent (Claude Code or Codex) at zero cost — FleetQ auto-detects them.
See Agents for setup.
1
Create an Agent
Go to /agents/create.
Fill in:
- Name: Competitor Monitor
- Role: Research Analyst
- Goal: Summarise competitor blog posts into concise bullet points
- Backstory: You are an expert technology journalist with 10 years of experience distilling complex content.
- Provider: Leave as Default (inherits team setting)
2
Create a Skill
Go to /skills/create.
- Name: Content Summariser
- Type: LLM
- Prompt:
Summarise the following article into exactly 3 bullet points. Be concise and focus on the key takeaway:\n\n{{ content }}
3
Assign the Skill to the Agent
Open the agent detail page (/agents → Competitor Monitor).
Under the Skills tab, click Add Skill and select Content Summariser.
Save. The agent now uses this skill when executing LLM calls.
4
Create a Project
Go to /projects/create.
- Name: Weekly Competitor Digest
- Type: Continuous
- Agent: Competitor Monitor
- Goal: Summarise the top 5 competitor blog posts published this week
- Schedule: Weekly — Monday at 08:00
5
Watch it run
On the project detail page, click Trigger Run to test immediately.
FleetQ creates an Experiment, runs it through the pipeline, and saves the output as an Artifact.
You can preview, download, or set up an outbound connector to deliver results via email or Slack.
Or use the API
All UI actions have API equivalents. Create your API token in Team Settings → API Tokens, then:
bash
curl -X POST https://fleetq.169.58.89.204.sslip.io/api/v1/experiments \
-H "Authorization: Bearer YOUR_TOKEN" \
-H "Content-Type: application/json" \
-d '{
"name": "Weekly Competitor Digest",
"agent_id": "YOUR_AGENT_ID",
"goal": "Summarise the top 5 competitor blog posts published this week"
}'
What next?
Experiments →
Learn about the 20-state pipeline, pausing, retrying, and artifacts.
Workflows →
Build visual DAG workflows with conditional branches and human tasks.
Signals →
Trigger workflows automatically from webhooks, RSS, or CRM events.
MCP Server →
Connect Claude Code or Cursor directly to FleetQ via 268+ MCP tools.