Neynar Inbox

PassAudited by ClawScan on May 10, 2026.

Overview

This is a coherent instruction-only email API skill, but it can create a mailbox, use a mailbox API key, send or delete emails, register webhooks, and set up polling, so users should keep those actions under clear control.

Use this skill only if you trust the Neynar Inbox API endpoint. Store the generated API key securely, require confirmation before sending or deleting emails, register webhooks only to trusted servers, and set any polling schedule with a clear stop condition.

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

An agent using this skill could send emails or delete mailbox data if directed or allowed to use the documented API calls.

Why it was flagged

The skill documents raw API calls for externally visible and mutating email actions. This is expected for an email API skill, but those actions can affect real recipients and mailbox contents.

Skill content
curl -X POST https://neynar-inbox-api.rish-68c.workers.dev/v1/emails ... "to": ["recipient@example.com"] ... | DELETE | /v1/emails/:id | Delete email |
Recommendation

Require user confirmation before sending, deleting, rotating keys, or registering webhooks, and limit actions to the intended mailbox and task.

What this means

Anyone with the API key may be able to send, read, search, delete, or manage emails for that mailbox.

Why it was flagged

The mailbox API key is the credential used to access the mailbox. This is disclosed and purpose-aligned, but possession of the key grants access to mailbox operations.

Skill content
"apiKey": "fi_abc123...", "warning": "Store the apiKey securely - it cannot be retrieved again."
Recommendation

Store the API key only in an approved secret store, avoid exposing it in logs or chat, and rotate or delete it if it may have been shared.

What this means

Email notifications could be sent to a configured server, so a wrong or untrusted webhook URL could expose mailbox activity.

Why it was flagged

The skill supports forwarding email event notifications to a webhook URL. This is disclosed and includes signature-verification guidance, but it creates an additional data-flow boundary.

Skill content
Register a webhook for real-time email notifications: ... -d '{"url": "https://your-server.com/webhook", "events": ["email.received"]}' ... Verify signatures via `X-Webhook-Signature` header
Recommendation

Use only trusted webhook endpoints, verify signatures, and avoid registering webhooks unless the user explicitly wants real-time notifications.

What this means

A scheduled agent could keep checking the mailbox and processing new email after the original task is done.

Why it was flagged

The skill recommends scheduled polling. This is coherent for receiving email replies, but it could continue beyond the immediate task if not bounded.

Skill content
RECOMMENDED: If you have heartbeat, cron, or scheduling capabilities, set up polling (every 30-60 seconds) right after creating your mailbox to catch replies.
Recommendation

Make polling opt-in, set a clear interval and end time, and provide an easy way to stop the schedule.