AIShield Security Scanner
Scans MCP servers for tool poisoning, prompt injection and supply chain risks.
Sollte ich dies verwenden
Qualität und Sicherheit
Befunde (4)
- LOWTool 'aishield_scan' description lacks action verbin aishield_scan
- LOWTool 'aishield_prompt_check' description lacks action verbin aishield_prompt_check
- LOWTool 'aishield_banned_words' description lacks action verbin aishield_banned_words
- LOWTool 'aishield_vertical_risk' description lacks action verbin aishield_vertical_risk
Basierend auf einer automatisierten Analyse der Tool-Definitionen und der Einhaltung des Protokolls.
Kontextkosten
Dies ist die ungefähre Anzahl der Tokens, die jedes Mal verbraucht werden, wenn die Tools des Servers in den Kontext eines Modells geladen werden. Höhere Werte verringern die Aufmerksamkeit, die für andere Aufgaben verfügbar ist.
Installieren
Installation mit einem Klick
Fügen Sie dies Ihrer Datei `claude_desktop_config.json` hinzu:
{
"mcpServers": {
"aishield": {
"command": "npx",
"args": [
"aishield-mcp-server"
]
}
}
}Ausführbare Pakete
4.3.0stdioRemote-Endpunkte
https://aishield.tools/api/v1/mcpstreamable-httpWas es kann
Tool-Inventar
Tools (10)
⚪aishield_scan(source_url, tool_type, name)
OWASP MCP Top 10 aligned security scan — 235 rules, 5-dimension scoring
Eingabe-Schema
{
"type": "object",
"properties": {
"source_url": {
"type": "string",
"description": "GitHub repo URL"
},
"tool_type": {
"type": "string",
"enum": [
"mcp",
"skill",
"gpt",
"prompt"
],
"default": "mcp"
},
"name": {
"type": "string",
"description": "Tool name"
}
},
"required": [
"source_url"
]
}🟢aishield_guardrail(source_url, auto_block)
Pre-install safety check — pass/block verdict
Eingabe-Schema
{
"type": "object",
"properties": {
"source_url": {
"type": "string",
"description": "GitHub repo URL"
},
"auto_block": {
"type": "boolean",
"default": true
}
},
"required": [
"source_url"
]
}🟢aishield_prompt_check(prompt)
Prompt injection detection — Chinese + English
Eingabe-Schema
{
"type": "object",
"properties": {
"prompt": {
"type": "string",
"description": "Prompt text to check (min 10 chars)"
}
},
"required": [
"prompt"
]
}⚪aishield_banned_words(text, platform)
Chinese banned words detection — 6 platform rules
Eingabe-Schema
{
"type": "object",
"properties": {
"text": {
"type": "string",
"description": "Text to check"
},
"platform": {
"type": "string",
"enum": [
"douyin",
"xiaohongshu",
"wechat",
"weibo",
"bilibili",
"kuaishou",
"all"
],
"default": "all"
}
},
"required": [
"text"
]
}🟢aishield_rug_pull(source_url)
Rug pull detection — check if a tool has removed security code or added suspicious changes in recent commits
Eingabe-Schema
{
"type": "object",
"properties": {
"source_url": {
"type": "string",
"description": "GitHub repo URL"
}
},
"required": [
"source_url"
]
}🟢aishield_handshake(source_url)
MCP handshake verification — analyze MCP config, detect npx auto-install, sensitive env vars, oversized tool descriptions, and attempt HTTP handshake
Eingabe-Schema
{
"type": "object",
"properties": {
"source_url": {
"type": "string",
"description": "GitHub repo URL"
}
},
"required": [
"source_url"
]
}⚪aishield_digest(configs, scan_result, source_url, max_findings)
Compact trust digest (aishield-digest/v1) — a few hundred bytes plus a content fingerprint, so an agent can answer 'can I trust this?' every turn without re-pulling the full report. Accepts {configs} (static analysis only), {scan_result}, or {source_url}. The returned `risk` is never lighter than the worst finding actually present (a config with high findings is never labelled 'safe'), and no plaintext credential is ever echoed back.
Eingabe-Schema
{
"type": "object",
"properties": {
"configs": {
"type": "object",
"description": "{path: file content} MCP client config map — static analysis, no command in the config is ever executed"
},
"scan_result": {
"type": "object",
"description": "An existing scan result to compress"
},
"source_url": {
"type": "string",
"description": "GitHub repo URL — return the current trust verdict as a digest"
},
"max_findings": {
"type": "integer",
"minimum": 0,
"maximum": 20,
"default": 3,
"description": "How many top findings to include"
}
}
}⚪aishield_vertical_risk(text, domain)
Vertical-industry risk scan — detect high-risk claims for finance/medical/gov sectors (unlicensed diagnosis, illegal medical device, financial over-promise, etc.)
Eingabe-Schema
{
"type": "object",
"properties": {
"text": {
"type": "string",
"description": "Text content to scan"
},
"domain": {
"type": "string",
"enum": [
"finance",
"medical",
"government"
],
"default": "finance",
"description": "Vertical domain"
}
},
"required": [
"text"
]
}🟢agent_register(agent_name, capabilities, owner)
Agent-First one-click onboarding — register as an Agent, get DID + API Key + quick start guide in a single call
Eingabe-Schema
{
"type": "object",
"properties": {
"agent_name": {
"type": "string",
"description": "Agent name (required)"
},
"capabilities": {
"type": "array",
"items": {
"type": "string"
},
"description": "Capability list, e.g. [\"scan\", \"monitor\"]"
},
"owner": {
"type": "string",
"description": "Owner identifier"
}
},
"required": [
"agent_name"
]
}⚪agent_quick_scan(tool_name, tool_description, source_url)
Agent-First quick scan — scan a tool by name and description, no source URL required
Eingabe-Schema
{
"type": "object",
"properties": {
"tool_name": {
"type": "string",
"description": "Tool name (required)"
},
"tool_description": {
"type": "string",
"description": "Tool description (required)"
},
"source_url": {
"type": "string",
"description": "Optional GitHub repo URL for deep scan"
}
},
"required": [
"tool_name",
"tool_description"
]
}Community
Nachweis