Skip to content

ASH MCP Streaming Results Guide

This guide explains how to use ASH's streaming functionality through the Model Context Protocol (MCP) server for real-time progress tracking during security scans.

Overview

The streaming functionality provides real-time progress updates during ASH security scans. Instead of waiting for a complete scan to finish, you can now:

  • Start scans in the background
  • Monitor progress in real-time
  • Get partial results as they become available
  • Manage multiple concurrent scans
  • Cancel running scans

This is especially useful for large codebases where scans can take several minutes to complete.

How It Works

ASH's MCP server uses an event-driven system to track scan progress. When you start a scan with progress tracking, ASH:

  1. Starts the scan in the background
  2. Returns a unique scan ID immediately
  3. Publishes progress events as the scan proceeds
  4. Allows you to check progress and get partial results
  5. Provides final results when the scan completes

Available Commands

When using ASH through an MCP-compatible AI assistant, you can use these streaming capabilities:

Starting a Streaming Scan

"Start a security scan with progress tracking on this directory"
"Begin a background security scan and show me updates as it runs"

This uses the scan_directory_with_progress tool, which returns immediately with a scan ID.

Monitoring Progress

"Show me the progress of my security scan"
"What's the current status of scan [scan-id]?"
"How many security issues have been found so far?"

This uses the get_scan_progress tool to show: - Overall progress percentage - Current scanning phase - Which scanner is currently running - Number of findings discovered so far - Estimated time remaining

Managing Multiple Scans

"Show me all my active security scans"
"List the status of all running scans"

This uses the list_active_scans tool to display all concurrent scans and their status.

Canceling Scans

"Cancel the security scan on the backend directory"
"Stop scan [scan-id]"

This uses the cancel_scan tool to stop a running scan and clean up resources.

Getting Final Results

"Get the final results for my completed security scan"
"Show me the detailed findings from scan [scan-id]"

This uses the get_scan_results tool to retrieve comprehensive results once a scan completes.

Progress Phases

Your security scan progresses through these phases:

  1. Initializing (0%): Setting up the scan and validating parameters
  2. Running (0-25%): Starting the scan execution
  3. Scanning (25-85%): Individual security scanners are running
  4. detect-secrets (finding hardcoded secrets)
  5. bandit (Python security issues)
  6. semgrep (code patterns and vulnerabilities)
  7. checkov (infrastructure security)
  8. And others based on your project type
  9. Generating Reports (85-95%): Creating output files and summaries
  10. Finished (100%): Scan completed successfully

Real-Time Updates

During the scanning phase, you'll see updates like:

  • "Running bandit scanner..." (Python security analysis)
  • "Running semgrep scanner..." (Pattern-based vulnerability detection)
  • "Running detect-secrets scanner..." (Credential scanning)
  • "Found 3 new security issues in current file..."
  • "Completed infrastructure security checks..."

Example Workflow

Here's a typical workflow using streaming scans:

  1. Start the scan: 

    "Start a security scan with progress tracking on my project"
    Response: "Started security scan with ID abc-123. I'll monitor the progress for you."

  2. Monitor progress: 

    "How is my security scan progressing?"
    Response: "Your scan is 45% complete. Currently running semgrep scanner. Found 2 security issues so far."

  3. Check multiple scans: 

    "Show me all my active scans"
    Response: "You have 2 active scans: Project A (75% complete), Project B (30% complete)."

  4. Get final results: 

    "My scan finished - show me the results"
    Response: "Scan completed! Found 5 security issues: 2 high severity, 3 medium severity. Here are the details..."

Benefits

For Large Projects

  • No need to wait for long scans to complete
  • See progress and early findings immediately
  • Can work on other tasks while scans run

For Multiple Projects

  • Run scans on different directories simultaneously
  • Track progress across all your projects
  • Prioritize which results to review first

For Development Workflow

  • Start scans when you begin code review
  • Get immediate feedback on critical issues
  • Cancel scans if you need to make changes

Configuration

The streaming functionality works with all your existing ASH configurations:

  • Severity thresholds: Set minimum severity levels
  • Custom scanners: Enable/disable specific security tools
  • Ignore patterns: Skip certain files or directories
  • Output formats: Choose how results are presented

Example requests: 

"Start a high-severity security scan with progress tracking"
"Scan this directory but ignore the test files, and show me progress"
"Run only the secrets scanner with streaming updates"

Troubleshooting

Scan Stuck or Taking Too Long

"Cancel my security scan and start a new one"
"Show me what scanner is currently running"

Multiple Scans Consuming Resources

"List all my active scans"
"Cancel all running scans except the most recent one"

Missing Progress Updates

"Check the status of scan [scan-id]"
"Is my security scan still running?"

Best Practices

1. Use for Large Codebases

Streaming is most beneficial for projects with many files or complex security configurations.

2. Monitor Critical Scans

For production or release-critical code, monitor progress to catch high-severity issues early.

3. Manage Resources

Don't run too many concurrent scans - they can consume significant CPU and memory.

4. Set Appropriate Thresholds

Use higher severity thresholds (HIGH or CRITICAL) for faster scans when you only need the most important issues.

Integration with Development Workflow

Code Review Process

  1. Start streaming scan when beginning code review
  2. Monitor for critical issues while reviewing
  3. Address high-priority security findings first

CI/CD Integration

  1. Start background scans early in the pipeline
  2. Monitor progress and fail fast on critical issues
  3. Generate detailed reports for successful builds

Daily Development

  1. Start scans on modified directories
  2. Work on other tasks while scans run
  3. Review findings when convenient

Next Steps

The streaming functionality makes security scanning a more integrated part of your development workflow, providing immediate feedback without blocking your productivity.