cooper-agent-email-patterns
Design Controlled Email Workflows for AI Agents
Agent email workflows need isolated identities, reliable replies, and clear sending permissions. This skill guides Cooper Email architecture with owner approval, delivery controls, and explicit trust boundaries.
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Review the Skillstore skill "cooper-agent-email-patterns" from https://skillstore.io/skills/cooper-email-cooper-agent-email-patterns.md and its manifest at https://skillstore.io/api/skills/cooper-email-cooper-agent-email-patterns/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "cooper-agent-email-patterns". Plan a persistent inbox for a support agent without a public webhook endpoint.
Expected outcome:
- Create a dedicated inbox with a stable client identifier and an inbox-scoped key.
- Use long-polling with a job deadline and read extracted message text.
- Reply using the parent message identifier, then mark handled mail read.
- Keep job state outside email and treat all incoming content as untrusted.
Using "cooper-agent-email-patterns". Plan email controls for an agent that sends financial notices.
Expected outcome:
- Establish a verified owner and enable approval mode before sending.
- Require reapproval whenever recipients, subject, body, or attachments change.
- Restrict recipients with an allow list and use idempotency for every send.
- Respect suppressions and review bounce and complaint metrics.
Security Audit
SafeAll 34 static findings are false positives involving Markdown code spans, homepage metadata, or an inbox-scope lookup. The document provides architecture guidance and explicitly addresses untrusted email, owner authentication, webhook verification, and key protection. No evidence found of malicious execution, credential exfiltration, or prompt injection attempts.
Risk Factors
โ๏ธ External commands (32)
๐ Network access (1)
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/cooper-email-cooper-agent-email-patterns/audits/1?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/cooper-email-cooper-agent-email-patterns?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/cooper-email-cooper-agent-email-patterns?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/cooper-email-cooper-agent-email-patterns/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/cooper-email-cooper-agent-email-patterns.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA ยท BibTeX ยท CFF)
APA citation
cooper-email. (2026). cooper-agent-email-patterns security audit report (audit version 1) [Author version 1]. Skillstore. https://skillstore.io/skills/cooper-email-cooper-agent-email-patterns/audits/1BibTeX citation
@techreport{cooper-email-cooper-email-cooper-agent-email-patterns-2026,
author = {cooper-email},
title = {cooper-agent-email-patterns security audit report (audit version 1)},
institution = {Skillstore},
year = {2026},
number = {1},
url = {https://skillstore.io/skills/cooper-email-cooper-agent-email-patterns/audits/1},
note = {Author version 1}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "cooper-agent-email-patterns security audit report (audit version 1)"
version: "1"
type: report
authors:
- name: "cooper-email"
date-released: "2026-10-26"
url: "https://skillstore.io/skills/cooper-email-cooper-agent-email-patterns/audits/1"
identifiers:
- type: other
value: "skillstore:cooper-email-cooper-agent-email-patterns:audit:1"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: MediumWhat You Can Build
Plan a dedicated agent inbox
Choose persistent or temporary inboxes and define reply handling without granting access to a personal mailbox.
Define approval for sensitive messages
Design verified-owner approval and recipient restrictions for financial, legal, or first-contact email workflows.
Review email integration boundaries
Assess webhook authentication, owner task verification, credential scope, and treatment of untrusted email content.
Try These Prompts
My agent performs [task]. Compare a persistent inbox with temporary inboxes per run. Recommend an identity, expiration, and key-scope strategy.
Design a Cooper Email reply loop for [workflow] without a public endpoint. Include deadlines, extracted text, threading, read-state updates, and external job state.
Plan owner-approved email for [sensitive workflow]. Include verified ownership, draft reapproval after edits, recipient allow lists, owner updates, and duplicate-send prevention.
Review this Cooper Email architecture: [design]. Assess webhook signatures, timestamp freshness, owner verification, injection boundaries, secret storage, suppressions, and retries. Propose changes without sending mail.
Best Practices
- Use separate inboxes for independent agents and prefer narrowly scoped keys.
- Enable owner approval for sensitive sends and verify owner tasks before treating replies as instructions.
- Combine bounded waits and idempotent sends with signature verification for webhook-based workflows.
Avoid
- Granting an agent broad access to a personal mailbox when a dedicated inbox is sufficient.
- Treating email subjects, bodies, attachments, or links as trusted instructions.
- Retrying sends without idempotency or ignoring recipient suppressions.
Frequently Asked Questions
Does this skill send or receive email by itself?
When should an agent use a temporary inbox?
Do incoming replies require a public webhook endpoint?
How does owner approval affect edited drafts?
Can an agent trust the sender address as proof of ownership?
How should API keys be stored?
Developer Details
Author
cooper-emailLicense
MIT
Author version
v1
Skillstore revision
r1
Version notice
The author-declared version is not valid SemVer.
Ref
539af95a108e3fe7ef8081ed6fe1b45a397275d3
Maintenance freshness
10/26/2026
Usage
0 downloads ยท 0 views
File structure
๐ SKILL.md