Resend CLI Skill

PassAudited by VirusTotal on May 11, 2026.

Overview

Type: OpenClaw Skill Name: resending-emails Version: 1.0.2 The skill bundle is a well-documented and functionally appropriate toolkit for enabling an AI agent to operate the Resend email service via its official CLI. It includes a Python helper script (scripts/resend_cli.py) designed to wrap CLI calls with deterministic flags, parse JSON/NDJSON output from both stdout and stderr, and provide diagnostic 'doctor' and 'lint-batch' utilities. The instructions in SKILL.md and the extensive reference documentation (references/agent-operating-model.md, etc.) focus on operational safety, such as recommending environment variables over command-line secrets and identifying known CLI coverage gaps. No evidence of malicious intent, data exfiltration, or harmful prompt injection was found; the use of subprocess execution is strictly aligned with the stated purpose of CLI automation.

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

An agent using this skill may send real emails, schedule broadcasts, create domains or webhooks, and manage Resend resources if given credentials.

Why it was flagged

The skill can guide an agent through live Resend account operations, including sending mail and mutating account resources. This matches the stated purpose, but users should recognize the operational impact.

Skill content
sending/listing/updating/cancelling emails, batch sends, domains and DNS, webhooks and local listeners, inbound receiving, contacts, topics, segments, broadcasts, templates, API keys, profiles
Recommendation

Use it only for Resend tasks you intend to run, prefer test/staging profiles first, and require explicit review before bulk sends, broadcasts, deletes, or account-wide changes.

What this means

If the agent has access to a broad Resend API key or production profile, mistakes could affect real customers or account configuration.

Why it was flagged

The skill needs Resend account credentials or a stored profile for live operations. This is expected for a Resend CLI integration, but it is sensitive account authority.

Skill content
Live operations require the official `resend` CLI plus a `RESEND_API_KEY` or stored profile.
Recommendation

Use least-privilege Resend API keys, domain-scoped tokens where possible, and explicit `--profile` selection for production versus staging.

What this means

Running installer scripts or global package installs affects the local environment and should be done only from trusted sources.

Why it was flagged

The skill documents user-run installation methods, including remote shell and PowerShell installers. These are common for CLIs and are not auto-executed by the skill, but they rely on trusted installation sources.

Skill content
`curl -fsSL https://resend.com/install.sh \| bash` ... `npm install -g resend-cli` ... `irm https://resend.com/install.ps1 \| iex`
Recommendation

Prefer official Resend installation channels, review installer commands before running them, and pin or verify CLI versions in CI where practical.

What this means

Webhook and inbound email payloads may contain sensitive message or recipient data and can be forwarded to local services during development.

Why it was flagged

The skill supports webhook and local listener flows that move Resend event payloads through public tunnel URLs and local endpoints. The behavior is disclosed and purpose-aligned, with verification guidance.

Skill content
starts a local HTTP server on `--port` ... creates a temporary webhook pointing at your public `--url` ... optionally forwards payloads to `--forward-to`
Recommendation

Use temporary tunnels carefully, verify webhook signatures, avoid exposing production payloads to untrusted local services, and stop listeners when finished.