# AgentAegis MCP Plugin Supply-Chain Trust Scanner

> AgentAegis MCP Plugin Supply-Chain Trust Scanner is a paid API for AI agents from agentaegis-mcp-production.up.railway.app, paid per call via x402, $5/call, status unknown (last checked 2026-09-15).

Performs a supply-chain security scan of an MCP server or agent skill (via git repo or code snippet) and returns a PROCEED/CAUTION/BLOCK verdict with categorized findings.

## Facts

- Endpoint: POST https://agentaegis-mcp-production.up.railway.app/x402/scan-mcp-plugin
- Price: $5/call
- Payment: x402
- Status: unknown
- Last checked: 2026-09-15
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/agentaegis-mcp-plugin-supply-chain-trust-scanner-fa23c23b
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_uGH855ZvJAJlA5KXMgfsV

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability agentaegis-mcp-plugin-supply-chain-trust-scanner-fa23c23b -d '<json body>'
```

Example prompt: Before I install this MCP server from https://github.com/someuser/some-mcp-plugin, run an AgentAegis supply-chain scan on it and tell me whether I should PROCEED, use with CAUTION, or BLOCK it — and flag any exfiltration, prompt injection, or dangerous capabilities you find.

## When to prefer this

Use this endpoint when an AI agent or developer needs to vet an MCP plugin or agent skill for supply-chain risks before installation or trust delegation — especially when you need a single authoritative PROCEED/CAUTION/BLOCK verdict covering exfiltration, prompt injection, dangerous capabilities, and obfuscation in one call. Prefer this over generic code scanners because it is purpose-built for the MCP/agent-skill threat model including hidden unicode and prompt-hijack detection.

## Known failure modes

- Git repo clone fails due to invalid URL or private repo without credentials — likely returns error with no verdict
- Code snippet too large or malformed — may return parse error
- Rate limit or payment verification failure for the $5 USDC x402 charge — returns 402 or payment error
- Scan timeout on very large repositories — may return partial results or timeout error
- Unsupported language or package manager — findings may be incomplete

## How this service works

AgentAegis scan_mcp_plugin — supply-chain trust scan of an MCP server or agent skill BEFORE you install/trust it. Clones a git repo (or takes a code snippet) and flags exfiltration (secrets/env to the network), prompt-injection sinks (hijack phrases + hidden unicode), dangerous capabilities (eval/shell/dynamic exec), npm install hooks, and obfuscation → one PROCEED/CAUTION/BLOCK verdict with findings.

## Output

Returns a verdict string (PROCEED, CAUTION, or BLOCK), a numeric trust score (0–100), a summary object with counts of exfiltration, prompt_injection, and dangerous_capabilities findings, and an array of human-readable reason strings explaining each flagged issue.

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/agentaegis-mcp-plugin-supply-chain-trust-scanner-fa23c23b/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from agentaegis-mcp-production.up.railway.app](https://www.zero.xyz/host/agentaegis-mcp-production.up.railway.app/llms.txt)