Pipeworx words

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 6:18 AM UTC
• Version: 1.0.0

Overview

This is a simple word-search skill that sends user-entered word queries to a remote MCP gateway, with no local code execution or persistence.

Before installing, understand that words or phrases you submit may be sent to gateway.pipeworx.io, apparently as an intermediary for Datamuse-powered results. Avoid submitting secrets, private drafts, personal data, or confidential internal terms unless you trust that gateway's handling of query data.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

Findings (1)

other

Note

• Location: SKILL.md:32

Finding

Undisclosed Transmission of User Queries Through a Third-Party Gateway

Content

View full analysis

Vulnerability Details

File Location: SKILL.md, lines 32–56
Vulnerability Type: Undisclosed third-party data disclosure
Risk Level: Low

The skill states that it is powered by Datamuse, but its documented requests transmit user-provided words and phrases to gateway.pipeworx.io, a separate third-party intermediary.

Vulnerable Code

bash

## Example: find rhymes for "code"

curl -X POST https://gateway.pipeworx.io/words/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"find_rhymes","arguments":{"word":"code","limit":10}}}'

## Example: words meaning "ocean" that start with "s"

curl -X POST https://gateway.pipeworx.io/words/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"find_words","arguments":{"meaning_like":"ocean","spelled_like":"s*"}}}'

{
  "mcpServers": {
    "words": {
      "url": "https://gateway.pipeworx.io/words/mcp"
    }
  }
}

Technical Analysis

The examples and MCP configuration direct all word-search requests to Pipeworx. This conflicts with the introductory representation that the toolkit is powered by the Datamuse API, because the documentation does not explain Pipeworx's role as an intermediary.

Queries are placed in JSON request bodies and can include arbitrary user-provided words or phrases. Pipeworx can consequently receive the submitted text and associated network metadata. If a user enters confidential text while seeking synonyms, related terms, or autocomplete results, that information may be disclosed outside the expected trust boundary.

No evidence indicates that credentials, local files, system data, or executable payloads are sent. The reviewed file also contains no local executable code.

Attack Path

1. A user or agent loads the documented MCP server configuration.
2. The user supplies a word or phrase to a word-search operation.
3. The client serializes that in ...[truncated 911 chars]

Remediation

View remediation

Remediation Suggestions

1. Clearly identify Pipeworx as the service receiving requests and explain whether it proxies Datamuse.
2. Document what query content and metadata are transmitted, retained, logged, or shared.
3. Obtain explicit user consent before transmitting potentially sensitive text to the third-party gateway.
4. Where feasible, use the official Datamuse endpoint directly to reduce unnecessary intermediaries.
5. Advise users not to submit secrets, credentials, personal data, or confidential text.
6. Provide a privacy policy and data-retention policy for the configured gateway.
7. If the gateway is required, apply contractual and technical controls covering encryption, retention limits, access control, deletion, and incident response.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

External Transmission

Medium

• Category: Data Exfiltration
• Confidence: 60% confidence

Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 37)May include surrounding context.

Example: find rhymes for "code"

bash

curl -X POST https://gateway.pipeworx.io/words/mcp \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"find_rhymes","arguments":{"word":"code","limit":10}}}'

Static analysis

No suspicious patterns detected.