Maverick Wordpress Mcp

PassAudited by VirusTotal on May 13, 2026.

Overview

Type: OpenClaw Skill Name: maverick-wordpress-mcp Version: 1.0.2 The skill is a legitimate Model Context Protocol (MCP) integration for WordPress.com. It uses a wrapper script (scripts/invoke.sh) and a vault seeder (scripts/init-mcporter.sh) to manage OAuth credentials provided via environment variables. The credential handling follows security best practices, such as using flock for concurrency, avoiding secrets in process arguments via jq environment variables, and storing tokens in a standard local directory (~/.mcporter/). All network activity is directed to the official WordPress.com API endpoint (public-api.wordpress.com).

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

The agent may be able to publish, edit, delete, moderate, or upload WordPress.com content if you authorize it.

Why it was flagged

The skill intentionally exposes mutation authority over WordPress.com content, but the same artifact discloses the impact and instructs confirmation before changes.

Skill content
Write-capable tools can change public or private WordPress.com content. Confirm clear user intent before creating, editing, publishing, unpublishing, deleting, moderating, or uploading content
Recommendation

Use it only for WordPress sites you want the agent to manage, and require explicit confirmation before any publishing, deletion, moderation, or upload action.

What this means

A refresh token in the local vault can allow continued WordPress.com access until it is revoked or rotated.

Why it was flagged

The wrapper persists OAuth access and refresh tokens into mcporter's local credential vault so mcporter can authenticate WordPress.com MCP calls.

Skill content
mcp_vault="${HOME}/.mcporter/credentials.json" ... tokens: {access_token: env.mcp_access, refresh_token: env.mcp_refresh, token_type: "Bearer"}
Recommendation

Protect the machine and the mcporter credential vault, avoid sharing these environment variables, and revoke the WordPress.com integration if you stop using the skill.

NoteHigh Confidence
ASI01: Agent Goal Hijack
What this means

WordPress.com's MCP server instructions may shape how the agent uses the tools during a session.

Why it was flagged

The skill tells the agent to rely on provider-published MCP instructions and schemas during use, which is expected for a thin pass-through MCP integration but gives remote instructions operational influence.

Skill content
The live server is the source of truth for what tools exist ... and any per-server instructions the server publishes. Treat this as the authoritative reference for the rest of the session.
Recommendation

Treat server-published instructions as tool documentation, while keeping system rules, user intent, and the skill's write-confirmation guidance higher priority.

What this means

Future changes to the mcporter package could affect how the skill runs.

Why it was flagged

The skill depends on installing the mcporter package by name; the provided install spec does not pin a version.

Skill content
node | package: mcporter | creates binaries: mcporter
Recommendation

Install mcporter from a trusted registry/source and prefer version pinning or package verification where the platform supports it.

What this means

WordPress.com receives the tool requests and content needed to perform the actions you ask for.

Why it was flagged

Authenticated MCP calls are sent to the declared WordPress.com hosted MCP endpoint over the configured HTTP transport.

Skill content
"baseUrl": "https://public-api.wordpress.com/wpcom/v2/mcp/v1", "transport": "http", "auth": "oauth"
Recommendation

Use the skill only for intended WordPress.com tasks and review WordPress.com's MCP/privacy documentation if you have data-sharing concerns.