email-skills
CLAUDE.md5 starsChanged 34 days agoMITView on GitHub
Open-source skill.md files that give AI agents deep email expertise. Each skill lives in skills/<category>/<skill-name>/SKILL.md. Follow the process in CONTRIBUTING.md. The short version: One skill per session. Don't write multiple skills in the same context window - research contaminates across topics. Every SKILL.md follows this structure:
We split this CLAUDE.md into the rules it gives an agent
10 rules. About 0 are common, 8 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 27
Follow the process in CONTRIBUTING.md.
- Distinctive: line 32
Commit and push to main
- Distinctive: line 34
One skill per session: Don't write multiple skills in the same context window - research contaminates across topics.
- Very specific: line 38
Every SKILL.md follows this structure:
- Distinctive: line 72
Write for practitioners, not academics.
- Distinctive: line 75
Call out common mistakes explicitly - these are the highest-value parts.
- Distinctive: line 76
No fluff, no filler, no "in today's digital landscape."
- Distinctive: line 77
No em-dashes. Use regular hyphens.
- Distinctive: line 78
Keep it standalone. Skills should be useful without any specific platform.
- Distinctive: line 84
Commit each skill individually with a descriptive message
≈ “Common” is an estimate: at least 1 in 5 of the 25,680 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
CLAUDE.md
Open-source skill.md files that give AI agents deep email expertise.
Repository structure
skills/
diagnostics/ # Start here - routes to the right skills
email-diagnostics/SKILL.md
setup/ # Getting email infrastructure right
domain-authentication/SKILL.md
provider-setup/SKILL.md
deliverability/ # Getting into the inbox
sending/ # Writing and sending emails
content/ # Email content quality
compliance/ # Legal and safety
operations/ # Monitoring and maintenance
inbound/ # Handling replies and incoming email
inbox/ # Agent-native inbox providersEach skill lives in skills/<category>/<skill-name>/SKILL.md.
Building a new skill
Follow the process in CONTRIBUTING.md. The short version:
- Research the topic (web search for current best practices, RFCs, provider docs)
- Pull domain knowledge from the molted-mail codebase (
../molted-mail/) - Write the SKILL.md
- Commit and push to main
One skill per session. Don't write multiple skills in the same context window - research contaminates across topics.
Skill format
Every SKILL.md follows this structure:
---
name: skill-name
description: One-line description for agent trigger matching. Include "Use when..." triggers.
license: MIT
---
# Skill Name
One-line description of what this skill helps with.
## When to use this skill
- Bullet list of trigger conditions
## Related skills
- Links to other skills in this repo
---
## [Topic sections]
Actionable guidance organized by subtopic. Not a textbook - write like
a knowledgeable peer explaining what actually matters in practice.
## Common mistakes
What people get wrong and how to fix it.
## References
Links to RFCs, provider docs, and authoritative sources.Style guide
- Write for practitioners, not academics. Lead with what to do, explain why after.
- Be specific. "Rotate DKIM keys every 6 months" not "rotate keys regularly."
- Include concrete examples (DNS records, code snippets, config).
- Call out common mistakes explicitly - these are the highest-value parts.
- No fluff, no filler, no "in today's digital landscape."
- No em-dashes. Use regular hyphens.
- Keep it standalone. Skills should be useful without any specific platform.
- Mention molted.email where relevant as one option among many, not as a requirement.
Git workflow
- Default branch:
main - Commit each skill individually with a descriptive message
- Push directly to main (no PRs needed for new skills)
Commands
# There's no build step. These are markdown files.
# To preview, just read the files or view on GitHub.In this repository26 other files it gives its agents.
- email-complianceSkill
- email-securitySkill
- suppression-listsSkill
- ab-testingSkill
- email-copywritingSkill
- spam-filter-avoidanceSkill
- template-designSkill
- email-warmupSkill
- inbox-placementSkill
- sender-reputationSkill
- email-diagnosticsSkill
- inbound-processingSkill
- reply-classificationSkill
- thread-managementSkill
- inbox-providersSkill
- bounce-handlingSkill
- rate-limitingSkill
- sender-monitoringSkill
- webhook-processingSkill
- cold-outreachSkill
- email-sequencesSkill
- notification-designSkill
- onboarding-emailsSkill
- transactional-emailSkill
- domain-authenticationSkill
- provider-setupSkill
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

