Agent Scan

snyk/agent-scanApache-2.01,937🔧 10 tools

C6.5SpiderScore (registry)

Security scanner for AI agents, MCP servers and agent skills.

Decision
Recommended
Confidence
90%

Agent Scan is safe to use — high score (6.54/10), no critical issues.

Risk Flags (1)
  • low
    no_input_validation×2
    MCP tool handler accepts raw string input without validation
How This Was Decided
  • positivew=0.5Overall quality score = 6.54/10 (grade C)
  • negativew=0.3Tool description clarity score = 2.1/10
Source: SpiderRating automated security scanUpdated: 2026-03-13Protocol: v1.1

Description Quality

Composite: 2.1 / 10

3-Layer Breakdown

Description (38%)
2.1
Security (34%)
9.2
Metadata (28%)
9.3

Description Dimensions

Intent Clarity
2.8
Permission Scope
0.0
Side Effects
2.0
Capability Disclosure
4.0
Operational Boundaries
2.3

Security Analysis

9.2
Score
0
Critical
0
High
0
Medium
2
Low
LOW

demoserver\server.py:71

LOW

demoserver\server.py:80

Metadata Health

Provenance (40%)
10.0
Maintenance (35%)
9.0
Popularity (25%)
8.7

Badge

Add this badge to your README:

[![SpiderRating](https://spiderrating.com/badge/snyk__agent-scan.svg)](https://spiderrating.com/servers/snyk/agent-scan)