Miro Developer App

ReviewAudited by ClawScan on May 10, 2026.

Overview

This appears to be a coherent Miro integration, but it requires Membrane/Miro authentication and can run actions that may change Miro workspace data.

Before installing, make sure you trust Membrane and the npm CLI package. During use, review OAuth scopes and confirm any action that creates, modifies, deletes, or manages Miro boards, users, or board content.

Findings (4)

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

Installing the CLI globally can run third-party package code and later versions may differ from the version reviewed here.

Why it was flagged

The skill asks the user to install a global npm package at the mutable @latest version. This is expected for a Membrane CLI integration, but it relies on the external npm package supply chain.

Skill content
npm install -g @membranehq/cli@latest
Recommendation

Install only if you trust Membrane and the npm package source; consider pinning or reviewing the CLI version in controlled environments.

What this means

The connected account may allow the agent to read or change Miro boards, members, and related data according to the granted scopes.

Why it was flagged

The integration requires delegated authentication and ongoing credential refresh for Membrane/Miro access. This is purpose-aligned, but it grants account-level authority to operate on Miro data.

Skill content
Membrane handles authentication and credentials refresh automatically ... The user completes authentication in the browser.
Recommendation

Review OAuth scopes during login, use the least-privileged account appropriate for the task, and revoke the Membrane/Miro connection when it is no longer needed.

What this means

A mistaken or overly broad action could create or modify Miro workspace content or other account data.

Why it was flagged

The skill enables generic action execution and direct Miro API proxying. That is consistent with an integration skill, but it can perform broad read/write operations if used without careful user direction.

Skill content
membrane action run <actionId> --connectionId=CONNECTION_ID --input '{"key": "value"}' --json ... When the available actions don't cover your use case, you can send requests directly to the Miro Developer App API
Recommendation

Confirm the exact action, endpoint, parameters, and expected effect before running create, update, delete, member-management, or direct proxy requests.

What this means

If provider-returned instructions were over-trusted, the agent could be steered into unintended follow-up actions.

Why it was flagged

The skill allows provider-returned connection state to include instructions for the agent. This can be useful setup guidance, but such text should not override the user's original goal or safety checks.

Skill content
`clientAction.agentInstructions` (optional) — instructions for the AI agent on how to proceed programmatically.
Recommendation

Treat returned agent instructions as contextual hints only, and require user confirmation for high-impact or unrelated actions.