CLI Reference

The WebroBot CLI (webrobot) is a Java-based command-line tool for managing all platform resources and running pipelines.

Installation

Download the latest release JAR and create a launcher script:

# Download CLI jar
curl -L https://github.com/WebRobot-Ltd/webrobot-cli/releases/latest/download/webrobot-cli.jar \
  -o ~/.local/share/webrobot-cli/webrobot-cli.jar

# Create launcher (requires Java 17+)
cat > ~/.local/bin/webrobot << 'EOF'
#!/usr/bin/env bash
java -jar "$HOME/.local/share/webrobot-cli/webrobot-cli.jar" "$@" 2> >(grep -v "^SLF4J:" >&2)
EOF
chmod +x ~/.local/bin/webrobot

Configuration

Create config.cfg in your working directory:

api_endpoint=https://api.webrobot.eu
apikey=your-api-key

Key	Description
api_endpoint	Base URL (default: https://api.webrobot.eu)
jwt / bearer / token	JWT auth — sends Authorization: Bearer <value>
apikey / apiKey	API key auth — sends X-API-Key header

Help: webrobot --help · webrobot <group> --help

---

project

Subcommand	Description	Key options
list	List all projects	—
add	Create project	-n name, -d description
get	Get project by ID	-i projectId
update	Update project	-i id, -n name, -d description
delete	Delete project	-i projectId
schedule-get	Get project schedule	-i projectId
schedule-set	Set cron schedule	-i projectId, -j jobId, -c cron expr, -e enabled

webrobot project list
webrobot project add -n "My Project" -d "Price monitoring"
webrobot project schedule-set -i 42 -j 7 -c "0 */6 * * *" -e true

---

category

Categories are containers for agents (not the same as projects).

Subcommand	Description	Key options
list	List all categories	—
get	Get by ID	-i categoryId
getbyname	Get by name	-n name
add	Create category	-n name, -d description
update	Update category	-i id, -n name
delete	Delete category	-i id

---

agent

Agents hold pipeline YAML definitions. They belong to a category (use -c).

Subcommand	Description	Key options
list	List agents in category	-c categoryId
get	Get agent	-c categoryId, -i agentId
getfromname	Get by name	-c categoryId, -n name
add	Create agent	-c categoryId, -n name, -d desc, -f pipelineYamlFile
update	Update agent	-c categoryId, -i agentId, -f pipelineYamlFile
delete	Delete agent	-i agentId

webrobot agent add -c 3 -n "price-monitor" -d "Monitoring pipeline" -f ./pipeline.yaml
webrobot agent update -c 3 -i 12 -f ./pipeline.yaml

---

job

Jobs link a project, an agent, and an input dataset.

Subcommand	Description	Key options
list	List jobs in project	-p projectId
add	Create job	-p projectId, -n name, -a agentId, -i inputDatasetId
update	Update job	-p projectId, -j jobId, -n name; optional: -a, -i, --enabled
delete	Delete job	-p projectId, -j jobId
execute	Run job	-p projectId, -j jobId; optional: --follow / -F
stop	Stop running job	-p projectId, -j jobId
logs	Get execution logs	-p projectId, -j jobId; optional: --tail, --pod-name

# Run and wait for completion
webrobot job execute -p 5 -j 12 --follow

# Stream logs
webrobot job logs -p 5 -j 12 --tail 200

--follow / -F: polls execution status every 5s, shows progress dots, exits with green ✓ on COMPLETED or red ✗ on failure.

---

pipeline

High-level commands for working with pipeline YAML manifests.

pipeline run

Apply a pipeline manifest and execute the job in one step:

webrobot pipeline run -f pipeline.yaml          # interactive confirm
webrobot pipeline run -f pipeline.yaml -y       # skip confirm
webrobot pipeline run -f pipeline.yaml --follow # run + wait

pipeline apply

Create/update project, agent, dataset, and job from a pipeline YAML (idempotent):

webrobot pipeline apply -f pipeline.yaml

Prints hints for webrobot job execute and webrobot pipeline run after success.

pipeline show

Show the current state of a pipeline, with ✓/? markers per stage (checked against stage catalog):

webrobot pipeline show -f pipeline.yaml

pipeline stages

webrobot pipeline stages list              # list all available stages (remote + local cache)
webrobot pipeline stages describe visit    # full detail of a stage: args, defaults, description
webrobot pipeline stages refresh           # invalidate cache, force-fetch from API

pipeline stages actions

webrobot pipeline stages actions list              # list browser automation actions
webrobot pipeline stages actions describe click    # detail + examples for an action

Available actions: click, type, scroll, wait, select, hover, navigate, submit, eval, none

---

dataset

Subcommand	Description	Key options
list	List datasets	optional: -t type, --indexed, -f format
get	Get dataset	-d datasetId
add	Create dataset	-n name, -d description
delete	Delete dataset	-d datasetId

---

package

Import/export projects as ZIP archives.

Subcommand	Description	Key options
exportproject	Export one project	-f output folder, -p projectId
exportall	Export all projects	-f output folder
importproject	Import project	-z zip file, -p projectId
importall	Import all	-z zip file

webrobot package exportproject -p 5 -f ./backups/
webrobot package importproject -p 5 -z ./backups/project-5.zip

---

Typical Workflow

# 1. Create a project
webrobot project add -n "price-comparison" -d "Competitor price monitoring"

# 2. Create a category and agent
webrobot category add -n "price-agents"
webrobot agent add -c <categoryId> -n "monitor" -f ./monitoring-pipeline.yaml

# 3. Create a dataset (trigger input)
webrobot dataset add -n "trigger-input"

# 4. Create a job linking them
webrobot job add -p <projectId> -n "monitor-run" -a <agentId> -i <datasetId>

# 5. Run the job and follow execution
webrobot job execute -p <projectId> -j <jobId> --follow

# Or: do steps 1-5 in one command from a YAML manifest
webrobot pipeline run -f pipeline.yaml --follow