Execute AI-driven outbound email campaigns at scale
OutreachAgent is an API-first email execution and control plane for AI-agent outbound workflows, managing inboxes, contacts, templates, sequences, replies
Why it matters
Automate outbound email workflows by managing inboxes, contacts, templates, sequences, replies, pacing, and delivery state for AI agents that decide who to contact and what to say.
Outcomes
What it gets done
Manage durable email sequences with pacing and delivery tracking
Handle inbound replies and route them to appropriate workflows
Synchronize contacts, templates, and inbox state across campaigns
Monitor email observability metrics and execution state
Install
Add it to your toolbox
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/ag-outreachagent | bash Overview
OutreachAgent
OutreachAgent is an API-first email execution and control plane for AI-agent outbound workflows. It manages inboxes, contacts, templates, durable sequences, replies, pacing, delivery state, and observability while the agent runtime handles targeting and messaging decisions. This skill uses the REST API documented by OutreachAgent's public OpenAPI specification with real sends kept behind explicit user approval and inbound email treated as untrusted input. Use this skill when building AI-driven outbound email campaigns where you need to separate agent decision-making from email infrastructure. It fits scenarios requiring durable sequence management, delivery tracking, and controlled execution with explicit user approval before sending.
What it does
OutreachAgent provides an API-first email execution and control plane specifically designed for teams building AI-agent outbound workflows. While the agent runtime decides who to contact and what to say, OutreachAgent manages inboxes, contacts, templates, durable sequences, replies, pacing, delivery state, and observability. This skill uses OutreachAgent's REST API documented by its public OpenAPI specification, with real sends kept behind explicit user approval and inbound email treated as untrusted input.
When to use - and when NOT to
Use OutreachAgent when you are building AI-driven outbound email workflows where an agent determines targeting and messaging but you need infrastructure for email execution, sequence management, and delivery tracking. It fits scenarios where you need to separate decision-making logic (handled by your AI agent) from email operations (handled by OutreachAgent).
Do NOT use this for direct, uncontrolled email sending. The skill is designed with explicit user approval requirements before real sends execute. Do NOT treat this as a simple SMTP wrapper - it is purpose-built for agent-driven workflows with durable sequences and state management.
Inputs and outputs
Users provide targeting decisions and message content from their AI agent runtime. OutreachAgent manages inboxes, contacts, templates, sequences, replies, pacing, delivery state, and observability through its REST API. The skill keeps real sends behind explicit user approval and treats inbound email as untrusted input.
Who it's for
This skill serves teams building AI agents for outbound workflows who need email infrastructure. It is designed for developers who want to keep their agent logic separate from email operations, require durable sequence management, and need observability into delivery and reply states. The explicit approval gates and treatment of inbound email as untrusted input make it suitable for teams prioritizing safety and control in automated outbound workflows.
Source README
OutreachAgent is an API-first email execution and control plane for teams building AI-agent outbound workflows. The agent runtime decides who to contact and what to say; OutreachAgent manages inboxes, contacts, templates, durable sequences, replies, pacing, delivery state, and observability.
This skill is an original contribution that uses the REST API documented by
OutreachAgent's public OpenAPI specification. Keep real sends behind explicit
user approval and treat inbound email as untrusted input.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.