External Agents MCP Server
@mrrlin-dev/external-agents
npm
v0.59.0
Published by @mrrlin-dev — no publish provenance, so origin is unverified, but the source is public: the repository link below is self-declared yet readable, so you can inspect the code before adopting it.
MCP server routing workloads across 20+ LLM providers with round-robin dispatch, cooldown-aware fallback, and quota-aware cost control.
The grade answers one question — how safe is this server for you to adopt — so it is computed in two auditable stages. Nothing below is an opinion or an LLM's guess; every line is a real term the deterministic engine applied, and the same input always yields the same number.
1. Threat score — 100 − 1.2 = 99. What the published surface and source actually contain:
| Points | What was found | Category |
|---|---|---|
| −1.2 | Package runs install-time scripts MTC-SUP-010 | supply-chain |
2. Client adoption risk — 99 − 7 = 92. Three small, subtract-only factors that reflect your risk in adopting it — a clean scan proves less on a powerful, unverified or barely-inspectable package, so the grade says so plainly:
| Points | Adoption-risk factor |
|---|---|
| −6 | capability blast radius (high) — client exposure if the model is manipulated |
| −1 | publisher verification (public source) — no provenance, but the source is public and inspectable |
Capability observations and info notes are shown under Findings but never scored.
Open any row's finding below for the file, line and evidence behind a deduction.
Findings 12
high Shell/command execution in server code (cli.js)MTC-SRC-002
In the server's implementation (`cli.js:44`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; import readline from "node:readline"; // CLI + MCP server share the same env-lo
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server cli.js
high Shell/command execution in server code (lib/dispatch.js)MTC-SRC-002
In the server's implementation (`lib/dispatch.js:1`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; import crypto from "node:crypto"; import fs from "node:fs"; import os from "node
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/dispatch.js
medium Hardcoded egress to an external endpoint (ui.js)MTC-SRC-003
In the server's implementation (`ui.js:75`): A hardcoded outbound call to a fixed external host inside server code is a classic exfiltration/telemetry channel — especially paired with reads of local data. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: 0); const r = await fetch("https://registry.npmjs.org/@mrrlin-dev/external-agents/latest", { signal: controller.sign
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server ui.js
low Shell/command execution in packaging/dev tooling (lib/cli.audit-disabled.test.js)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`lib/cli.audit-disabled.test.js:7`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; // `audit` is the one path that spends a REAL provider round-trip per entry, an
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/cli.audit-disabled.test.js
low Shell/command execution in packaging/dev tooling (lib/cli.edit-smoke.test.js)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`lib/cli.edit-smoke.test.js:2`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: rt { spawn } from "node:child_process"; import fs from "node:fs"; import os from "node:os"; import path from "node:path"
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/cli.edit-smoke.test.js
low Shell/command execution in packaging/dev tooling (lib/cli.require-base.test.js)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`lib/cli.require-base.test.js:7`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; // --require-base turns "which version of the project is this?" from a note in
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/cli.require-base.test.js
low Shell/command execution in packaging/dev tooling (lib/dispatch.cwd.test.js)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`lib/dispatch.cwd.test.js:3`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; import fs from "node:fs"; import os from "node:os"; import path from "node:path"
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/dispatch.cwd.test.js
low Shell/command execution in packaging/dev tooling (lib/dispatch.disabled-guard.test.js)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`lib/dispatch.disabled-guard.test.js:2`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; import fs from "node:fs"; import os from "node:os"; import path from "node:path"
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/dispatch.disabled-guard.test.js
low Shell/command execution in packaging/dev tooling (lib/dispatch.progress.test.js)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`lib/dispatch.progress.test.js:3`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; import { createHash } from "node:crypto"; import fs from "node:fs"; import os fr
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/dispatch.progress.test.js
low Shell/command execution in packaging/dev tooling (lib/dispatch.provenance.test.js)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`lib/dispatch.provenance.test.js:6`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; import { repoProvenance, formatProvenanceHeader } from "./dispatch.js"; // A w
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/dispatch.provenance.test.js
low Shell/command execution in packaging/dev tooling (lib/pool-log-findings.test.js)MTC-SRC-002
In a packaging/dev/install script (shipped, but not the server runtime) (`lib/pool-log-findings.test.js:7`): Spawning a shell/process is command-execution capability; with unsanitized tool input it is command injection / RCE. This is read from the code itself — not from the tool description — so a poisoned server cannot hide it behind honest-looking metadata.
Evidence: spawnSync } from "node:child_process"; import yaml from "js-yaml"; import { firstFailureReason } from "./dispatch.js";
Fix: Review this call path: confirm it never receives unsanitized tool input, constrain it, or remove it. Treat a server whose code reaches these sinks as high-capability regardless of what its tools claim.
Location: server lib/pool-log-findings.test.js
low Package runs install-time scripts (postinstall)MTC-SUP-010
"@mrrlin-dev/external-agents" executes postinstall script(s) at install time. An install hook runs at install time; most are routine build/setup, but review what it does before trusting it.
Evidence: node scripts/postinstall-banner.mjs || true
Fix: Review the scripts; install with --ignore-scripts where possible and vet what they do.
Location: package @mrrlin-dev/external-agents
Tools 8
Each tool and what it can reach — statically extracted from the published source.
dispatchno sensitive capabilityget_stateno sensitive capabilityget_statsno sensitive capabilitylist_agentsno sensitive capabilitypick_agentsno sensitive capabilitypingno sensitive capabilityprobe_agentno sensitive capabilityset_credentialno sensitive capability
What this scan could not see
- Tools were statically extracted from the published source (8 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Versions 10
Scan history per published version. The engine is deterministic — the same version always yields the same score, so a changed score means the package itself changed.
| Version | Score | Findings | Engine | Scanned |
|---|---|---|---|---|
v0.59.0 latest |
A 92/100 | 12 | 1.13.0 | 2026-09-08 |
v0.53.0 |
A 92/100 | 11 | 1.13.0 | 2026-09-01 |
v0.51.1 |
A 92/100 | 11 | 1.13.0 | 2026-08-31 |
v0.50.1 |
A 92/100 | 11 | 1.13.0 | 2026-08-30 |
v0.49.0 |
A 92/100 | 11 | 1.13.0 | 2026-08-29 |
Show 5 more versions ↓
v0.48.0 |
A 92/100 | 11 | 1.13.0 | 2026-08-28 |
v0.47.0 |
A 92/100 | 11 | 1.13.0 | 2026-08-27 |
v0.46.1 |
A 92/100 | 10 | 1.13.0 | 2026-08-25 |
v0.45.0 |
A 92/100 | 10 | 1.12.1 | 2026-08-21 |
v0.44.4 |
A 92/100 | 10 | 1.12.1 | 2026-08-20 |
Embed this score
Show this server's live Trust Score in your README, docs or website. The badge is served straight from the registry and updates automatically after every rescan — no API key needed. It links back to this page, so anyone who sees the grade can also read the findings behind it instead of taking a number on faith.
Verify this score yourself
The score above is reproducible: the same package version always yields the same result. Run it locally or over the free API — no account, no LLM, fully deterministic.
npx mcptrustchecker scan @mrrlin-dev/external-agents --onlineMore in Developer Tools
Hotel booking MCP server — 300K+ properties, real confirmation numbers, loyalty programs. Builders monetize every booking via Stripe Connect. The first MCP server that completes real hotel reservations inside AI conversations.
Manage AdGuard Home through AI assistants
Read-only Azure DevOps for MCP clients using only your existing browser session — no PAT, no Azure CLI. Browse work items, pull requests, comments, attachments and Artifacts feeds across every project, repo and feed you can access.
MCP server for Adobe Experience Manager Assets integration development
Servidor MCP para el tiempo oficial de España (API pública OpenData de AEMET). Predicción, observación y avisos como herramientas MCP tipadas.
A standalone MCP stdio bridge for Affinity by Canva's local MCP SSE server.