RelayShield Agentic Attack Surface

by relayshieldadminco.huggingface.relayshieldadmin/relayshield-agentic-attack-surface

  • Remote endpoint
  • 13 tools

Uptime

Checked  · every day

100%from 4 days of checks

Up without downtime on 4 of 4 days with data.

  • No downtime
  • Degraded
  • Down over 5 minutes
  • No data
  • Not checked yet

Response time

Median 1196 ms, slowest 5% 1395 ms for the handshake and the tool list, over 30 days.

1500 ms750 ms0

How far each check got

  • Reached the address 100%
  • Completed the handshake 100%
  • Listed its tools 100%

13 tools at the last answer.

Every day
DayResultChecksMedian
3 Nov (today)No downtime11395 ms
2 NovNo downtime1of 11160 ms
1 NovNo downtime1of 11183 ms
31 OctNo downtime11213 ms

A failure counts as down time only when a second check five minutes later fails too. How we check

Connect to the publisher's own endpoint

Add to ~/.config/zed/settings.json:

{
  "context_servers": {
    "relayshield-agentic-attack-surface": {
      "url": "https://relayshieldadmin-relayshield-agentic-attack-surface.hf.space/gradio_api/mcp/sse"
    }
  }
}

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

For agents: the install prompt

Every step above as one Markdown text, for an assistant installing it for you.

# Install the MCP server RelayShield Agentic Attack Surface (`co.huggingface.relayshieldadmin/relayshield-agentic-attack-surface`)

Its page: https://plug.directory/servers/co.huggingface.relayshieldadmin/relayshield-agentic-attack-surface

It runs on the publisher's servers; the client connects to its URL.

Use the section for the person's client.

## Claude Desktop

Paste into Settings › Connectors › Add custom connector:

```text
https://relayshieldadmin-relayshield-agentic-attack-surface.hf.space/gradio_api/mcp/sse
```

## Claude Code

Run in a terminal:

```sh
claude mcp add --transport sse relayshield-agentic-attack-surface https://relayshieldadmin-relayshield-agentic-attack-surface.hf.space/gradio_api/mcp/sse
```

## Cursor, Devin Desktop, LM Studio, Gemini CLI, oterm, ECA

Merge into the place for the person's client:

- Cursor: ~/.cursor/mcp.json
- Devin Desktop: ~/.config/devin/mcp_config.json
- LM Studio: mcp.json (Program › Install › Edit mcp.json)
- Gemini CLI: ~/.gemini/settings.json
- oterm: config.json in oterm's data folder (~/Library/Application Support/oterm on macOS, ~/.local/share/oterm on Linux, %APPDATA%\oterm on Windows)
- ECA: ~/.config/eca/config.json (global) or .eca/config.json in the project

```json
{
  "mcpServers": {
    "relayshield-agentic-attack-surface": {
      "url": "https://relayshieldadmin-relayshield-agentic-attack-surface.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## VS Code

Merge into .vscode/mcp.json:

```json
{
  "servers": {
    "relayshield-agentic-attack-surface": {
      "type": "sse",
      "url": "https://relayshieldadmin-relayshield-agentic-attack-surface.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## Zed

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Cline, AnythingLLM, Augment Code

Merge into the place for the person's client:

- Cline: cline_mcp_settings.json (MCP Servers › Configure › Configure MCP Servers)
- AnythingLLM: anythingllm_mcp_servers.json in the AnythingLLM storage plugins directory
- Augment Code: Augment Settings Panel – MCP section – Import from JSON (VS Code extension)

```json
{
  "mcpServers": {
    "relayshield-agentic-attack-surface": {
      "type": "sse",
      "url": "https://relayshieldadmin-relayshield-agentic-attack-surface.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## Continue

Merge into ~/.continue/config.yaml:

```yaml
mcpServers:
  - name: "relayshield-agentic-attack-surface"
    type: sse
    url: "https://relayshieldadmin-relayshield-agentic-attack-surface.hf.space/gradio_api/mcp/sse"
```

## Goose

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Codex

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## mcp-client-cli, TUUI, BoltAI, Bob Shell, Kiro, Kiro CLI

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Agenvoy

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Amp

Merge into ~/.config/amp/settings.json (user) or .amp/settings.json (workspace):

```json
{
  "amp.mcpServers": {
    "relayshield-agentic-attack-surface": {
      "url": "https://relayshieldadmin-relayshield-agentic-attack-surface.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

Then restart the client, or reload its MCP servers, and ask it to list the server's tools to confirm it started.

Tools 13

  • relayshield_agentic_attack_surface_check_agent_risk_summary Composite risk check -- one call instead of four. Runs breach exposure, LLM credential exposure, and tech-stack CVE checks together and returns a single fused risk verdict, so a calling agent doesn't have to make multiple tool calls and reason about how to combine their results itself. Returns: A structured result combining all sub-checks that were run, with the, highest severity found across them driving the overall outcome.
  • relayshield_agentic_attack_surface_check_bulk_identity_risk Hierarchical org + AI-agent-identity risk scoring for a domain and up to 5 agent/service-account identities. A critically-exposed agent automatically elevates the organizational risk rating. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_llm_credential_exposure Check a domain for exposed LLM/AI provider API keys (OpenAI, Anthropic, Google, Groq, xAI, Replicate) in criminal stealer logs -- LLMjacking, a fast-growing threat where a leaked key becomes a live, uncapped billing liability rather than just a data exposure. Real incidents have run from tens of thousands of dollars per day to a $500K single-month bill from one leaked, unthrottled key. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_mcp_server_risk Check an MCP server URL for typosquat risk, presence in RelayShield's criminal IOC corpus, and domain-registration age. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_nhi_exposure Check a domain, or up to 10 vendor domains, for exposed non-human- identity (NHI) credentials — API keys, service-account tokens, personal access tokens, and other machine identities found in criminal stealer logs. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_oauth_watchlist Check an email for OAuth-connected-app exposure: apps matched against breach history, plus stolen OAuth/session tokens found in criminal stealer logs. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_prompt_injection_breach Check an email address for credential exposure sourced specifically from prompt-injection attacks against AI agents, distinct from ordinary phishing/malware-sourced breaches. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_secret_scan Check a domain, or up to 5 vendor domains, for secrets and credentials exposed in public GitHub repositories. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_server_status Check which RelayShield tools and data sources are currently live on this MCP server -- a lightweight introspection call for an agent (or a human debugging) to confirm connectivity and see what's available before making a real check. Requires no API key. Returns: A structured result listing available tools and a live upstream connectivity check.
  • relayshield_agentic_attack_surface_check_session_risk Check an email for active or reusable stolen session material (cookies, tokens) found in criminal stealer logs — exposure that can bypass MFA/authentication entirely, not just a password. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_supply_chain Check up to 10 vendor domains for breach and infostealer exposure — third-party and supply-chain risk inherited through vendor relationships. Returns: A structured result: outcome/recommended_action plus evidence and reason codes.
  • relayshield_agentic_attack_surface_check_tech_stack_cve Check a declared AI agent framework / tech stack for CISA KEV or high-EPSS CVEs actively being exploited. Covers AI agent orchestration frameworks (LangChain, CrewAI, AutoGPT, Flowise, n8n self-hosted) and common companion infrastructure (Nacos, MinIO). Returns: A structured result: outcome/recommended_action plus evidence and reason codes.

From Hugging Face Spaces. Report a problem