Systems Lab

Agent skill

cold-email-deliverability

Audits and guides cold email deliverability — plain text sending, SPF/DKIM/DMARC, ESP setup, recipient filters, and SEG awareness.

slowingSelf-containedActs undeclared831 words

Filed under Outbound email.

From bcharleson/claude-code-cold-email-skills · 3 skills · 6 · pushed 2026-07-10

What it does when it runs

Audits and guides cold email deliverability — plain text sending, SPF/DKIM/DMARC, ESP setup, recipient filters, and SEG awareness. Use when reviewing sending infrastructure, inbox placement, DNS records, email authentication, spam filters, Proofpoint, Mimecast, Barracuda, or deliverability before a campaign.

Read from the skill and the 1 file bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
None found.
Hosts it reaches
No third-party host appears in the skill or its bundled files.
Tool permissions it declares
No allowed-tools in the frontmatter. It does act, so it runs under whatever permissions your session already grants.
Actions present in the files
shell

Ask about cold-email-deliverability

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah
git clone --depth 1 --filter=blob:none --sparse https://github.com/bcharleson/claude-code-cold-email-skills.git /tmp/claude-code-cold-email-skills
git -C /tmp/claude-code-cold-email-skills sparse-checkout set ".claude/skills/cold-email-deliverability"
mkdir -p ~/.claude/skills/cold-email-deliverability
cp -R "/tmp/claude-code-cold-email-skills/.claude/skills/cold-email-deliverability/." ~/.claude/skills/cold-email-deliverability/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Reproduced in full from bcharleson/claude-code-cold-email-skills/blob/9e1d7f735caf9c1de42da275f45abdfc149bdc33/.claude/skills/cold-email-deliverability/SKILL.md, which is licensed MIT (repository). 831 words, 21 headings.

Built for Claude Code (and compatible agent harnesses). Drop into .claude/skills/.

Cold Email Deliverability

Deliverability is always top of mind. Legitimate cold outreach shares the same filtering environment as spam. Infrastructure and message format must earn inbox placement before copy or offers matter.

Agent Responsibilities

When deliverability is in scope:

  1. Audit before send — Check DNS auth, message format, and recipient-environment risks.
  2. Flag blockers immediately — Missing SPF, DKIM, or DMARC is a hard stop. Do not recommend sending until resolved.
  3. Default to safest format — Plain text, no links, no images, unless the user explicitly chooses otherwise.

Message Format Rules

Default (best practice)

  • Plain text only — No HTML templates, no rich formatting.
  • No links — URLs trigger marketing/spam heuristics. Omit unless intentionally doing so with eyes open on risk.
  • No images — Same rationale as links.
  • No attachments in initial cold touches — increases filter scrutiny.

When links or assets are intentional

If the user explicitly wants links (calendar, deck, video):

  • Acknowledge deliverability tradeoff.
  • Prefer sending the asset after a positive reply.
  • Ensure domain reputation, DNS auth, and landing pages are solid (see SEG section).

Sending-Side Infrastructure

Every sending domain or inbox — Google Workspace, Microsoft 365, or custom SMTP — must have proper DNS authentication.

Required DNS records

RecordPurposeAgent action
SPFAuthorizes which servers may send for the domainVerify present and includes only legitimate send sources
DKIMCryptographic signature proving message integrityVerify signing is enabled and DNS publishes the public key
DMARCPolicy for failed SPF/DKIM; reportingVerify policy exists (start with p=none for monitoring if needed, tighten over time)

If any of SPF, DKIM, or DMARC is missing or misconfigured → flag immediately. Deliverability will suffer; do not proceed as if send-ready.

Additional sending hygiene

  • Use a dedicated outreach domain or subdomain when possible — protects primary brand domain.
  • Warm new domains/inboxes gradually — volume spikes hurt reputation.
  • Align From domain with DKIM signing domain.
  • Keep bounce and complaint rates low; remove invalid addresses promptly.
  • Match reverse DNS (PTR) where applicable on custom SMTP.

Recipient-Side Environment

Prospects receive mail through varied stacks. ICP often determines the mix.

Common recipient ESPs

  • Google (Gmail / Workspace) — Strong filtering, engagement signals matter.
  • Microsoft (Outlook / M365) — Similar; sensitive to authentication failures and bulk patterns.
  • Third-party / custom SMTP — Variable rules; authentication failures are penalized heavily.

The "handshake" between sender and receiver depends on authentication, reputation, content signals, and engagement history.

SEGs (Secure Email Gateways)

Corporate inboxes often sit behind SEGs that add layers beyond native ESP filtering:

  • Proofpoint
  • Barracuda
  • Mimecast
  • Cisco IronPort / others

Implications for outreach:

  • Authentication must be flawless — SEGs are stricter than consumer Gmail.
  • Domain reputation and domain age matter more.
  • Domain masking — Sending domain should look legitimate and aligned with the business (not deceptive; professional and consistent).
  • Landing pages — If links are used later, pages must be trustworthy: valid SSL, clear business identity, no redirect chains, no mismatch between email domain and link domain.
  • Plain-text-first cold touches reduce SEG triage into "marketing" buckets.

When ICP is enterprise or regulated (healthcare, finance, government), assume SEG presence and apply stricter infrastructure standards.

Pre-Send Checklist

Copy this checklist and work through it:

Deliverability audit:
- [ ] SPF record present and correct for sending source
- [ ] DKIM enabled and DNS key published
- [ ] DMARC record present with sensible policy
- [ ] Sending domain aligned with From address
- [ ] Message is plain text (unless user opted into HTML/links)
- [ ] No links or images in initial touch (unless intentional)
- [ ] Volume appropriate for domain/inbox age
- [ ] Recipient environment considered (consumer vs corporate/SEG)
- [ ] Landing pages valid if links will be used in follow-ups

Audit Output Format

When reviewing a user's setup or draft campaign, respond with:

## Deliverability Assessment

### Status: [READY | NEEDS WORK | BLOCKED]

### DNS Authentication
- SPF: [✅ / ⚠️ / ❌] — [detail]
- DKIM: [✅ / ⚠️ / ❌] — [detail]
- DMARC: [✅ / ⚠️ / ❌] — [detail]

### Message Format
- [plain text / HTML / links / images] — [pass or flag]

### Recipient Risk
- ICP environment: [e.g. corporate + likely SEG]
- Notes: [specific risks]

### Required Actions (ordered)
1. ...

Why This Matters

Good deliverability plus relevant messaging lets legitimate offers reach the right person. One inbox placement can start a conversation with real impact — but only if authentication, format, and trust signals are in place first.

Reference

For DNS check commands and SEG notes, see checklist.md.

Files bundled with it

These load only when the skill asks for them, so they cost nothing until it runs.

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what cold-email-deliverability does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.