Shellward
by jnmetacodecommunityv0.5.11
55
8
ShellWard is an AI agent security middleware that protects against prompt injection, data exfiltration, and dangerous command execution by intercepting tool calls at runtime.
From the registry: AI agent security: 7 MCP tools for injection detection, PII scanning, command safety, DLP.
Install & configure
Prerequisites
- requiredClone repositoryMust clone repo first, then run via tsx
https://github.com/jnMetaCode/shellward$ git clone https://github.com/jnMetaCode/shellward && cd shellward && npm install
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `shellward` MCP server into my current AI client (that's you).
Required prerequisites (do these first if not already done):
- **Clone repository** — Must clone repo first, then run via tsx Run: `git clone https://github.com/jnMetaCode/shellward && cd shellward && npm install` (https://github.com/jnMetaCode/shellward)
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["tsx","<path-to-shellward>/src/mcp-server.ts"]
- optional environment variables:
- `SHELLWARD_MODE`: enforce or audit mode (default: enforce) (example: `enforce`)
Note: AI agent security middleware. 8-layer defense: prompt injection detection, PII scanning, dangerous command blocking, DLP data flow control. 7 MCP tools. Zero dependencies. Also available as npm SDK.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: AI agent security middleware. 8-layer defense: prompt injection detection, PII scanning, dangerous command blocking, DLP data flow control. 7 MCP tools. Zero dependencies. Also available as npm SDK.
Tools (7)
check_command
Check if a shell command is safe (rm -rf, reverse shell, fork bomb...)
check_injection
Detect prompt injection in text (32+ rules, zh+en)
scan_data
Scan for PII & sensitive data (CN ID/phone/bank, API keys, SSN...)
check_path
Check if file path operation is safe (.env, .ssh, credentials...)
check_tool
Check if tool name is allowed (blocks payment/transfer tools)
check_response
Audit AI response for canary leaks & PII exposure
security_status
Get current security config & active layers
Environment variables
SHELLWARD_MODErequiredSets the mode of operation, either 'enforce' or 'audit'.SHELLWARD_LOCALErequiredSets the locale for the middleware, options include 'auto', 'zh', or 'en'.SHELLWARD_THRESHOLDrequiredSets the risk score threshold for injection detection, ranging from 0 to 100.Packages
shellward
npmv0.5.11transport: stdio
Details
- Language
- TypeScript
- License
- Apache-2.0
- Released
- Mar 23, 2026
- Last commit
- Apr 1, 2026
- Registry name
- io.github.jnMetaCode/shellward