Back to skill

Security audit

Zouroboros Swarm Gate

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local task-complexity classifier and its behavior matches that purpose, with only ordinary supply-chain and routing-heuristic caveats.

Before installing, prefer a pinned package version or lockfile-managed install instead of copying unpinned npx examples. Treat its output as routing advice, since common words like 'just' or 'quick' can lower the orchestration score.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (17)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied code chunk is only a tsup build config and does not implement or expose the described classifier behavior. While it also makes no API calls, that alone does not align it with the declared primary purpose. The actual code's purpose is packaging/building source files, which is materially different from a task classifier that evaluates signals to decide on multi-agent orchestration.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

The README instructs users to run the package via npx without pinning a specific version. This creates a supply-chain risk because the command will resolve to the latest published package version at execution time, so a compromised maintainer account, malicious release, or dependency confusion event could cause unreviewed code to run on a user's machine.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

This example again uses npx without a version constraint, which means users may execute whatever package version is current at the time they run the command. In a security-sensitive tooling ecosystem, that increases exposure to malicious or broken releases and makes behavior non-reproducible.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The documented override triggers use extremely common words like just, quick, and simple, which are likely to appear in ordinary user requests unrelated to routing intent. That makes the classifier easy to manipulate accidentally or deliberately, potentially downgrading tasks that actually require multi-agent review or forcing orchestration when not intended.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

The documentation instructs users to run an unpinned package via npx, which will fetch and execute whatever version is current at invocation time. This creates a supply-chain risk: a compromised publisher account, malicious new release, or unexpected breaking change could lead to arbitrary code execution on the user's machine.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
94% confidence
Finding

This example again uses unversioned npx, causing users to execute the latest published package without review or reproducibility guarantees. In the skill context, users may copy-paste this directly, so any malicious or compromised future release could execute attacker-controlled code immediately.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
96% confidence
Finding

The help text instructs users to run the package via npx @zouroboros/swarm-gate without pinning a specific version. That means executions may fetch the latest published package at runtime, creating a supply-chain risk if a future release is compromised, replaced, or unexpectedly changes behavior. In a CLI/security-adjacent workflow, encouraging unpinned remote execution is meaningfully risky because users may copy-paste the command directly.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
96% confidence
Finding

This second usage example also recommends npx @zouroboros/swarm-gate --json without an exact version pin, exposing users to the same runtime package-substitution and unexpected-update risk. Because npx may download and execute code on demand, any compromise of the package or its publication path can translate directly into code execution on the user's machine.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

The help text instructs users to run the package via npx @zouroboros/swarm-gate without a pinned version. npx will fetch the latest published package at execution time, so a compromised publisher account, malicious update, or dependency confusion-style package replacement could cause users to execute unexpected code. Because this is a CLI intended to be run directly, the risk is more meaningful than a passive documentation issue.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

This second usage example repeats the unpinned npx @zouroboros/swarm-gate pattern, creating the same supply-chain execution risk. In JSON mode, users are still executing whatever version is current at the time of invocation, which may differ from the reviewed or expected code.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Known Vulnerable Dependency: esbuild==0.27.7 — 1 advisory(ies): GHSA-g7r4-m6w7-qqqr (esbuild allows arbitrary file read when running the development server on Window)

Low
Category
Supply Chain
Confidence
93% confidence
Finding

The lockfile pins esbuild to 0.27.7, which is reported as affected by GHSA-g7r4-m6w7-qqqr. In this package the dependency is marked as a devDependency, so the issue is most likely only reachable during local development or build workflows, but if the esbuild development server is used on Windows it could allow unintended arbitrary file reads.

Content

No source excerpt is available for this finding.

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
40% confidence
Finding

Dependencies lack version pinning, allowing potential malicious package updates. Consider pinning versions.

Content

Scanner excerpt · package.json (reported line 49)May include surrounding context.

json
"prepublishOnly": "npm run build"
  },
  "devDependencies": {
    "tsup": "^8.0.0",
    "typescript": "^5.4.0"
  },
  "engines": {

Unpinned Dependencies

Low
Category
Supply Chain
Confidence
40% confidence
Finding

Dependencies lack version pinning, allowing potential malicious package updates. Consider pinning versions.

Content

Scanner excerpt · package.json (reported line 50)May include surrounding context.

json
},
  "devDependencies": {
    "tsup": "^8.0.0",
    "typescript": "^5.4.0"
  },
  "engines": {
    "node": ">=22.0.0"