Skip to main content
The stop command gracefully shuts down all Shannon containers. Optionally, it can remove all data including Docker volumes, workspaces, and cached images.

​
Basic Usage

# Stop containers (preserves data)
./shannon stop

# Stop and remove all data
./shannon stop CLEAN=true

​
Behavior

​
Default Stop (Data Preserved)

./shannon stop
What happens:
  1. Gracefully stops all running containers:
    • temporal (Temporal server)
    • worker (Shannon worker)
    • router (claude-code-router, if using ROUTER=true)
  2. Preserves all data:
    • Workflow state in Temporal database
    • Audit logs in ./audit-logs/
    • Deliverables in ./repos/<repo>/deliverables/
    • Docker volumes (Temporal database, Elasticsearch)
  3. Allows resume from last state
Use when:
  • Temporarily stopping Shannon
  • Switching between projects
  • System maintenance
  • Preserving workflow history for later resume

​
Clean Stop (Remove All Data)

./shannon stop CLEAN=true
​
CLEAN
boolean
default:"false"
Remove all data including Docker volumesDeletes:
  • Temporal database (all workflow history)
  • Elasticsearch data (log storage)
  • Docker volumes
  • Container state
Preserves:
  • Audit logs in ./audit-logs/
  • Deliverables in ./repos/<repo>/deliverables/
  • Source repositories in ./repos/
  • Configuration files
CLEAN=true permanently deletes all Temporal workflow history. You cannot resume workflows after a clean stop.Audit logs and deliverables in the filesystem are preserved.
Use when:
  • Complete reset needed
  • Troubleshooting Temporal issues
  • Freeing disk space
  • Starting fresh environment

​
Examples

​
Standard Workflow

# Start pentest
./shannon start URL=https://example.com REPO=my-repo WORKSPACE=audit

# Stop for the day (preserves state)
./shannon stop

# Resume next day
./shannon start URL=https://example.com REPO=my-repo WORKSPACE=audit
# Continues from where it left off

​
Clean Restart

# Stop and clean everything
./shannon stop CLEAN=true

# Start fresh
./shannon start URL=https://example.com REPO=my-repo

​
Development Testing

# Run quick test
./shannon start URL=https://example.com REPO=my-repo PIPELINE_TESTING=true

# Clean up after test
./shannon stop CLEAN=true

​
What Gets Preserved

​
Default Stop (./shannon stop)

Data TypeLocationPreserved?
Workflow stateTemporal DB✅ Yes
Workflow historyTemporal DB✅ Yes
Audit logs./audit-logs/✅ Yes
Deliverables./repos/<repo>/deliverables/✅ Yes
Docker volumesDocker✅ Yes
Configuration.env, ./configs/✅ Yes
Source repositories./repos/✅ Yes

​
Clean Stop (./shannon stop CLEAN=true)

Data TypeLocationPreserved?
Workflow stateTemporal DB❌ Deleted
Workflow historyTemporal DB❌ Deleted
Audit logs./audit-logs/✅ Yes
Deliverables./repos/<repo>/deliverables/✅ Yes
Docker volumesDocker❌ Deleted
Configuration.env, ./configs/✅ Yes
Source repositories./repos/✅ Yes

​
Container Management

​
Stopped Containers

After ./shannon stop, containers remain but are stopped: 
# View stopped containers
docker compose ps -a

# Container status: Exited
Benefits:
  • Fast restart with ./shannon start
  • No re-download of images
  • Preserved volume mounts

​
Removed Containers

After ./shannon stop CLEAN=true, containers are removed: 
# View containers (none exist)
docker compose ps -a
# No Shannon containers
Impact:
  • Next ./shannon start recreates containers
  • Volumes deleted (Temporal DB, Elasticsearch)
  • Images remain cached (no re-download needed)

​
Resume After Stop

​
Resume After Default Stop

# Stop containers
./shannon stop

# Resume workflow (same command as original start)
./shannon start URL=https://example.com REPO=my-repo WORKSPACE=audit

# Output:
# Starting Shannon containers...
# Temporal is ready!
# Resuming workspace: audit
# Skipping completed agents: pre-recon, recon
# Continuing from: injection-vuln
How it works:
  1. Containers restart with preserved volumes
  2. Temporal DB contains complete workflow history
  3. Audit logs indicate completed agents
  4. Workflow resumes from last checkpoint

​
No Resume After Clean Stop

# Clean stop (deletes Temporal DB)
./shannon stop CLEAN=true

# Attempt resume
./shannon start URL=https://example.com REPO=my-repo WORKSPACE=audit

# Output:
# Starting Shannon containers...
# Temporal is ready!
# Starting new workflow: audit_shannon-1234567890
# Note: Previous workflow history deleted
What happens:
  • Temporal DB is empty (no workflow history)
  • Audit logs still exist but not linked to active workflow
  • New workflow starts from beginning
  • Can still access old audit logs and deliverables manually
Even after CLEAN=true, you can manually inspect previous audit logs in ./audit-logs/<workspace-id>/ and deliverables in ./repos/<repo>/deliverables/.

​
Stopping Router

If using multi-model routing (ROUTER=true), the router container is also stopped: 
# Router runs with --profile router
./shannon start URL=https://example.com REPO=my-repo ROUTER=true

# Stop all containers including router
./shannon stop
# or
./shannon stop CLEAN=true
Both commands stop the router container using --profile router.

​
Manual Container Management

For advanced users, you can manage containers directly:

​
View Running Containers

docker compose ps

​
Stop Specific Container

# Stop worker only
docker compose stop worker

# Stop router only
docker compose --profile router stop router

​
Remove Containers Without Volumes

# Remove containers but keep volumes
docker compose down

​
Remove Volumes Manually

# Remove all volumes
docker compose down -v

# Remove specific volume
docker volume rm shannon_temporal-db

​
Disk Space Management

Shannon containers and volumes can consume significant disk space:

​
Check Disk Usage

# View Docker disk usage
docker system df

# View Shannon volumes
docker volume ls | grep shannon

​
Free Disk Space

# Clean stop (removes volumes)
./shannon stop CLEAN=true

# Remove unused Docker data
docker system prune -a --volumes
docker system prune -a --volumes removes all unused Docker data, not just Shannon. Use with caution if you have other Docker projects.

​
Troubleshooting

​
Containers Won’t Stop

# Force stop containers
docker compose down --timeout 1

# Force remove if still stuck
docker compose rm -f

​
Volumes Won’t Delete

# Ensure containers are stopped
./shannon stop

# Manually remove volumes
docker volume rm shannon_temporal-db shannon_elasticsearch

# Or force clean
./shannon stop CLEAN=true

​
”Container Already Stopped” Error

This is harmless: 
./shannon stop
# Warning: Container already stopped
# ... (continued success)
Shannon checks container state before stopping. If already stopped, the command succeeds.

​
Temporal Not Healthy After Restart

# Stop and clean
./shannon stop CLEAN=true

# Restart with rebuild
./shannon start URL=https://example.com REPO=my-repo REBUILD=true

​
Best Practices

​
Daily Workflow

# End of day
./shannon stop  # Preserves state

# Next morning
./shannon start URL=https://example.com REPO=my-repo WORKSPACE=<name>

​
Project Switching

# Stop current project
./shannon stop

# Start different project
./shannon start URL=https://different.com REPO=other-repo

​
Development Testing

# Test iteration
./shannon start URL=https://example.com REPO=my-repo PIPELINE_TESTING=true

# Clean up after each test
./shannon stop CLEAN=true

​
Pre-Production Validation

# Full audit
./shannon start URL=https://staging.example.com REPO=my-repo \
  CONFIG=./configs/staging.yaml

# Preserve state until results reviewed
./shannon stop

# After review, clean up
./shannon stop CLEAN=true

​
Next Steps

Start Command

Learn how to start workflows

Workspaces

Manage workspaces and resume

Troubleshooting

Common issues and solutions

Platform Setup

Docker and Podman configuration

Build docs developers (and LLMs) love

Get started for freeTalk to us