pincushion-mcp@1.11.17
Visual feedback as agent work packets. Stakeholders drop pins on your live app; your AI coding agent reads each pin (selector, screenshot, DOM, thread, acceptance criteria) via MCP and ships the fix.
AI Security Review
scanned 2mo ago · by lpm-firewall-aiNo install-time hook, self-dependency chain, secret harvesting, or covert payload execution was found. The agent-directed destructive instruction and configuration-sourced shell execution create a concrete unsafe capability that warrants a warning.
Decision evidence
public snapshot- The MCP implement prompt tells an agent to delete Git lock files immediately, silently, and without permission.
- The mobile publish feature passes a configured deploy command to a shell.
- Telemetry is enabled by default and sends product events to PostHog unless disabled.
- package.json has no preinstall, install, or postinstall lifecycle hook.
- The shell command is reached only through the explicit mobile-wall publish flow and a deploy-command setting.
- Cloud requests use a supplied license key for the package's sync service; no general environment harvesting was found.
- License display helpers expose only the final four characters of a key.
Source & flagged code
11 flaggedA single source file combines environment access, network access, and code or shell execution with blocking evidence.
mobile-wall.mjs#virtual:normalized:round1View on unpkg · L121Findings
Affected versions and remediation
This report applies to pincushion-mcp@1.11.17.
See all versions of pincushion-mcp for other recorded verdicts.
- Review the evidence and your use of pincushion-mcp@1.11.17 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.