feat: add --dry-run option for email and broadcast commands

[Shubhdeep12]

• To validate without API calls.
• Currently added for emails send and broadcast create commands as they are write-heavy.

---

Summary by cubic

Adds a --dry-run mode to emails send and broadcasts create to validate payloads and print the exact request JSON without calling the API. Output is { "dryRun": true, "request": { ... } } and works interactively without network calls or an API key.

• New Features
  • resend emails send --dry-run: validates and prints { "dryRun": true, "request": { ... } }; attachments summarized as filename and byteLength; respects --json; skips verified-domain lookup in interactive mode.
  • resend broadcasts create --dry-run: validates and prints the create payload (includes send/scheduledAt when set); respects --json; in interactive mode, prompts for segment/topic IDs instead of fetching lists.
  • Help text and docs updated (README.md, skills/resend-cli/SKILL.md); other write commands remain unchanged (no dry-run yet).

^{Written for commit 1d1f5ff. Summary will update on new commits.}

[cubic-dev-ai]

2 issues found across 6 files

Prompt for AI agents (unresolved issues)

Check if these issues are valid — if so, understand the root cause of each and fix them. If appropriate, use sub-agents to investigate and fix each issue separately.


<file name="src/commands/emails/send.ts">

<violation number="1" location="src/commands/emails/send.ts:78">
P2: Dry-run still performs API-related work before early return (requireClient and fetchVerifiedDomains), which can call the API and require credentials despite the “without calling the API” promise.</violation>
</file>

<file name="src/commands/broadcasts/create.ts">

<violation number="1" location="src/commands/broadcasts/create.ts:56">
P2: `--dry-run` is documented as making no API calls, but the command still performs API-backed lookups (domains/topic/segment pickers) before the dry-run check, so interactive dry runs can still hit the API.</violation>
</file>

_{Reply with feedback, questions, or to request a fix. Tag @cubic-dev-ai to re-run a review, or fix all with cubic.}

[cubic-dev-ai]

No issues found across 6 files

[felipefreitag]

@Shubhdeep12 Thanks for adding this.
The interactive mode naturally calls the API, so it seems to me that the --dry-run flag makes sense only for non-interactive mode, what do you think?

[felipefreitag]

All other keys are in camelCase, I think this one should be too

[felipefreitag]

All other keys are in camelCase, I think this one should be too

[Shubhdeep12]

@Shubhdeep12 Thanks for adding this. The interactive mode naturally calls the API, so it seems to me that the --dry-run flag makes sense only for non-interactive mode, what do you think?

hey @felipefreitag I’d suggest keeping interactive --dry-run too, so the flag always means no calls to Resend.
we’d just skip the listing APIs and use text prompts where we’d normally fetch lists.

[felipefreitag]

See discussion about interactive mode