security-scan
AgentShield を使用して、Claude Code の設定（.claude/ ディレクトリ）のセキュリティ脆弱性、設定ミス、インジェクションリスクをスキャンします。CLAUDE.md、settings.json、MCP サーバー、フック、エージェント定義をチェックします。
82
73%
Does it follow best practices?
Impact
97%
2.25xAverage score across 3 eval scenarios
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./docs/ja-JP/skills/security-scan/SKILL.mdEvaluation results
91%
51%Claude Code Configuration Security Audit
Security audit with structured output
Uses ecc-agentshield
0%
100%
Checks install first
0%
100%
Targets .claude/ directory
0%
100%
JSON output format
0%
100%
Output file saved
100%
100%
Grade/score present
100%
100%
Critical findings identified
100%
100%
Severity interpretation
100%
100%
Min-severity filter used
0%
0%
No alternative tool
0%
100%
100%
69%Automated Security Gate for Claude Code Configurations
CI/CD GitHub Actions integration
Correct action reference
0%
100%
path parameter set
0%
100%
min-severity parameter set
0%
100%
min-severity value is medium or stricter
0%
100%
fail-on-findings enabled
0%
100%
Triggers on pull_request
100%
100%
Proper YAML action syntax
0%
100%
No hardcoded secrets
100%
100%
Workflow file location
100%
100%
Documentation included
80%
100%
100%
41%Bootstrapping a Secure Claude Code Project Configuration
Secure configuration initialization
Uses init command
0%
100%
Does not manually craft config
0%
100%
settings.json has scoped permissions
100%
100%
settings.json has deny list
100%
100%
CLAUDE.md created
100%
100%
mcp.json placeholder created
100%
100%
Post-init scan run
50%
100%
Scan uses ecc-agentshield
0%
100%
Scan result saved
100%
100%
Setup script produced
100%
100%
- Repository
- affaan-m/everything-claude-code
- Commit
5df943e
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.