Index of every page on this site: llms.txt
All skills
Run it yourself

Operate Resend through its official CLI for authorised sends, hosted templates, domains, contacts, broadcasts, inbound email, webhooks, or Automations. Use for explicit Resend terminal/CI operations; use resend-api for application SDK code, not this skill for generic email or inbox tasks.

Use this Skill: https://skilld.dev/gh/tristanmanchester/agent-skills/resend-cli

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈75 tokens for metadata: the name and description. ≈1.3k when used: this file. ≈1.4k more on demand in 2 files.

Description uses 3.8% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈302 characters, or 3.8%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Resend CLI operations

Use the maintained CLI directly. Its commands output replaces the old duplicated command catalogue, task router, and Python wrapper. Preserve task-specific intent and verification, not stale copies of every flag.

Discover and scope

resend --version
resend --json commands
resend emails send --help
resend --json -q doctor

doctor can inspect the account; it is not a purely offline version check. Use an already authorised connector instead when that is the available execution surface. Do not install, log in, change profiles, or broaden credentials just to run a read-only inspection. Check effective account/profile and token scope without printing keys. Prefer a secret-store environment value or stored profile over --api-key in arguments. Record the CLI version and help used.

For bounded automation, pass every required argument, --json -q, and capture stdout, stderr, and exit status separately. Current machine output is success JSON on stdout and errors on stderr; warnings may also occur. Exit zero and a valid expected result are both needed. Stream listeners are not bounded JSON commands. Do not concatenate output channels and pretend the result is one JSON document. Inspect installed help when a flag or output shape differs.

Choose the operation

Intent Surface Verification
One transactional email emails send Persist ID; inspect delivery result later
Distinct transactional messages emails batch Freeze payload, map returned IDs, reconcile failures
Hosted content templates and emails send --template Published version, variables, sender defaults
Marketing campaign broadcasts, segments, topics Exact audience/consent, preview, send status
Sender/receiving setup domains Returned DNS records and verified capabilities
Received message emails receiving Fetch full content/attachments after verified event
Event-driven email sequence automations, events Disabled draft, approved activation, run results
Credential/event delivery setup api-keys, webhooks Least privilege, secret storage, test delivery

Exact flags come from resend commands and subcommand help, not this routing table. Do not declare an operation unsupported just because a cached catalogue lacks it. Use the API/SDK only after confirming a relevant CLI gap.

Hosted-template send

Current emails send supports --template and --var. It is not necessary to fall back to REST just to send a hosted template.

resend --json -q emails send \
  --template "$TEMPLATE_ID" --to delivered@resend.dev \
  --var CUSTOMER_NAME=Taylor --idempotency-key "$SEND_KEY" --dry-run

Resolve an actual published template and its variable definitions first. Supply sender/subject when not defined by the template; verify overrides. Current CLI --var values are strings: use an appropriate typed SDK/API payload when numeric variable types matter. Do not combine template with raw body flags; the inspected CLI also rejects attachments with templates. CLI restrictions need not imply the same restriction on every API surface.

The dry run validates local input, not template existence, publication, permissions, recipient authorisation, or delivery. Review its payload; remove --dry-run only for the authorised send, retaining the same content and operation identity.

Mutation and retry discipline

Confirm actual recipients, account, content, attachments, schedule/timezone, and scope before sending. Drafting or previewing is not permission to send. Sending a broadcast or firing a custom event may fan out to many recipients.

Persist a unique idempotency key and exact request before email/batch submission. Retry the same operation with that key only within the provider's documented window (currently 24 hours). New content is a new operation. After the window, reconcile rather than replay; do not extend email idempotency guarantees to other endpoints. A timeout may mean accepted-but-unacknowledged, not unsent.

Native --dry-run exists for specific commands (currently email send and broadcast create), not every mutation. Inspect help; never emulate a preview by making a live call. For unsupported previews, render the intended payload and recipients locally without invoking the mutation.

Read operational checks for batching, domains, subscriptions, receiving, streams, and Automations. Report IDs, confirmed state, unknown outcomes, and next evidence needed. Accepted is not delivered, and recorded delivery is not proof of inbox placement or human reading.

Sources and evaluation

Reviewed against the official CLI, send source, and API docs on 2026-09-13. Evaluation cases: evals/scenarios.json. They require agent trace review; no live account access or send is implied by their presence.

Source: SKILL.md on GitHub

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

1 warning1mo4 checks · Risk SAFE
  • Gen Agent Trust Hub1mo

    The skill is a legitimate and well-structured tool for managing the official Resend CLI. It includes documentation, assets, and a Python helper script designed to facilitate safe command execution and environment diagnostics. The remote installation scripts mentioned are official vendor resources for the Resend service.

  • Socket1mo

    No alerts

  • Snyk1mo

    Risk: MEDIUM · 2 issues

  • ZeroLeaks6mo

    Score: 93/100 · 2 sections analyzed

Signed by skilld at 3323bc9. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub 2 weeks ago.

Steadyupdated last month
Other metadata
compatibility
Requires the official resend CLI and authorised Resend access for live operations. Discover the installed command surface before execution; no bundled Python wrapper is required.
metadata
{
  "version": "4.0.0",
  "reviewed": "2026-09-13",
  "source": "https://github.com/resend/resend-cli"
}

README badge

README badge for tristanmanchester/agent-skills/resend-cli