IDE Integration

Deep integration with your editor for a seamless security workflow. Every component is designed to surface findings where you already work — no context switching required.

Security Dashboard

The dashboard provides a visual summary of your security posture. Open it via the Command Palette: Open Security Dashboard.

The dashboard displays:

Severity breakdown — Count of findings by Critical, High, Medium, and Low
Recent scan results — Timestamp, duration, and finding counts from the last scan
Quick actions — One-click buttons to scan, export, or configure analyzers

The dashboard updates automatically after every scan.

Sidebar Views

The Security Scanner adds a dedicated panel to the Activity Bar (the shield icon). It contains eight tree views organized by function:

View	What it shows
MCP Security Findings	Findings organized hierarchically: Config → Server → Tool → Finding. Expand any node to see details, severity, and analyzer source.
Skills Security Findings	Findings organized by skill, with each finding showing its category, severity, file path, and line number.
MCP Configurations	All discovered MCP configuration files with their scan status (scanned, pending, error). Includes both global and workspace configs.
Skills	All discovered skills from Cursor, Claude, Codex, Antigravity, workspace, and custom paths. Shows scan status and source.
Allowlist	Trusted servers, tools, and skills that are excluded from scan results. Add items directly from findings via right-click.
Scan History	Past scan results with timestamps, finding counts, and duration. Compare any two scans to see new, resolved, and unchanged findings.
Watchdog Findings	Protected files grouped by preset with status indicators (unchanged, modified, missing). Actions include View Diff, Restore, Accept Change, and Re-Snapshot.
Settings	Quick access to toggle analyzers, notifications, and scan scope without leaving the sidebar.

Sidebar view showing MCP Configurations with scan status indicators

Sidebar view showing Skills security findings with severity counts

Interacting with findings:

Click a finding to open its detail panel with full description, recommendation, and metadata
Right-click a finding for actions: Add to Allowlist, Remove Server, Report False Positive, Rescan
Click a skill finding with a file path to jump directly to the relevant line in your editor

Detailed view of a high-severity finding showing threat type, analyzer, description, and actionable recommendations

CodeLens and Inline Decorations

CodeLens — In MCP configuration files (JSON/JSONC), a clickable CodeLens annotation appears above each server definition. It shows the scan status and finding count for that server. Click it to run a targeted scan on that specific config.

Inline decorations — After a scan, status indicators appear directly in MCP config files next to server and tool entries:

Green checkmark for safe items
Yellow/red indicators for items with findings

Both CodeLens and inline decorations can be toggled via settings:

mcp-scanner.ui.codeLens (default: on)
mcp-scanner.ui.inlineDecorations (default: on)

Skill file annotations — When a skill scan finds issues, the scanner adds:

CodeLens above the relevant section in the skill file
Inline decorations highlighting the specific lines
Diagnostics entries in the Problems panel

Problems Panel and Quick Fixes

All findings from both MCP and skill scans are surfaced as diagnostics in VS Code's built-in Problems panel (Cmd/Ctrl+Shift+M). This means findings appear alongside your linter errors, type errors, and other diagnostics in a single unified view.

Each diagnostic includes:

Severity mapped to VS Code's Error/Warning/Information levels
Message with the finding title and threat type
Location pointing to the relevant file and line

Quick fixes (the lightbulb icon) are available on each diagnostic:

Add to Allowlist — Suppress this finding for the identified server, tool, or skill
Remove Server — Remove the flagged server from the MCP config
Report False Positive — Flag the finding for review

False positive reporting dialog showing finding details and explanation field

Status Bar

A persistent status bar item in the bottom-left corner shows the scanner's current state at a glance:

State	Meaning
Initializing	Extension is loading, setting up Python environment
Ready	Idle and ready to scan
Scanning	A scan is currently in progress
Safe	Last scan completed with no findings
Issues Found	Last scan found one or more findings (shows count)
Error	Something went wrong — check the output panel
Setup Required	First run — open the Setup Wizard to configure

Click the status bar item to run Scan All (MCP + Skills).

Report Export

Export scan results for sharing, compliance, or CI/CD integration. Run Export Scan Report from the Command Palette and choose a format:

Format	Best for
JSON	Machine-readable output, CI/CD pipelines, programmatic analysis
Markdown	Human-readable reports, documentation, team sharing
CSV	Spreadsheet analysis, tracking over time, management reporting

The report includes all findings from the most recent scan, with full details: severity, analyzer, threat type, description, and recommendation.

Scan Comparison

Compare the current scan with a previous one to track your security posture over time. Run Compare with Previous Scan from the Command Palette.

Scan History sidebar view showing past scan results and duration

The comparison shows:

New findings — Threats that appeared since the previous scan
Resolved findings — Issues that are no longer present
Unchanged findings — Persistent issues that still need attention

This is particularly useful after updating MCP server versions, modifying skills, or adding items to the allowlist.