Crowdin

WarnAudited by ClawScan on May 10, 2026.

Overview

This appears to be a real Crowdin integration, but it gives the agent broad authenticated Crowdin API access, including direct write/delete requests, without clear guardrails.

Review this skill before installing. It is appropriate if you want an agent to operate Crowdin through Membrane, but only connect accounts with suitable permissions and require explicit confirmation before the agent creates, updates, or deletes Crowdin resources.

Findings (5)

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

If used carelessly, the agent could make broad changes to Crowdin projects, files, users, webhooks, or other account data.

Why it was flagged

This authorizes direct authenticated API calls outside predefined actions, which can bypass safer scoped workflows.

Skill content
When the available actions don't cover your use case, you can send requests directly to the Crowdin API through Membrane's proxy... injects the correct authentication headers
Recommendation

Require explicit user approval for non-read operations, prefer scoped Crowdin actions, and restrict proxy use to user-specified endpoints and methods.

What this means

Crowdin content or configuration could be changed or deleted if the agent chooses an unsafe endpoint or method.

Why it was flagged

The documented proxy supports mutating and destructive HTTP methods, including DELETE, without any stated confirmation or containment rule.

Skill content
`-X, --method` | HTTP method (GET, POST, PUT, PATCH, DELETE). Defaults to GET
Recommendation

Add a clear policy that DELETE, POST, PUT, and PATCH require the user's specific confirmation and a preview of the target resource.

What this means

The Membrane connection may retain access to Crowdin until revoked.

Why it was flagged

Credential delegation and refresh are expected for a Crowdin integration, but they grant ongoing account access through Membrane.

Skill content
Membrane handles authentication and credentials refresh automatically
Recommendation

Use the least-privileged Crowdin account or token available and revoke the Membrane connection when no longer needed.

What this means

The installed CLI version could change over time and was not statically reviewed here.

Why it was flagged

The skill relies on installing the latest global CLI package from npm, which is purpose-aligned but unpinned and not part of the reviewed artifact set.

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

Install from a trusted npm source, consider pinning a known CLI version, and keep the CLI updated through normal package-management controls.

What this means

Remote setup instructions could influence agent behavior if treated as authoritative.

Why it was flagged

The skill may expose remote service-provided instructions to the agent during connection setup; this is useful but should not override the user's request or safety limits.

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

Treat returned agent instructions as operational hints only, and keep the user's request and safety rules as the controlling authority.