Skip to main content

​
Founder OS Template

Founder OS is built for founders and agency operators who spend their days on go-to-market activities: managing a pipeline, creating content, monitoring social channels, and handling communications. Your agent operates as a GTM co-pilot with a critical safety feature — all external communications require human approval before sending.

​
What’s Included

​
Skills

SkillWhat It Does
githubMonitors repos for product updates to surface in content
gogGame tracking — because founders need downtime too
apple-remindersTask management and follow-up reminders

​
Integrations

IntegrationModeDetails
TelegramLivePrimary communication channel — agent messages you here
LinkedInDraft-onlyAgent drafts posts and comments, you approve before publish
TwitterDraft-onlyAgent drafts tweets and threads, you approve before publish
GmailDraft-onlyAgent drafts email responses, you approve before send

​
Dashboard Focus

SectionWhat It Shows
GTM PipelineLeads, deals, outreach status, conversion tracking
Content CalendarScheduled posts, drafts pending approval, published content
LinkedIn MonitoringRelevant conversations, engagement opportunities, DM summaries
Twitter MonitoringMentions, relevant threads, trending topics in your space
CommsEmail drafts, Telegram conversations, pending approvals

​
Heartbeat

Founder OS checks in on a regular cycle to:
  1. Scan LinkedIn and Twitter for relevant conversations and engagement opportunities
  2. Review pipeline status and surface deals that need attention
  3. Check for emails requiring response
  4. Prepare content drafts based on the content calendar
  5. Send a Telegram summary only when something needs your action

​
Draft-Only Mode: The Safety Guardrail

All external communications are draft-only by default. Your agent will never post to LinkedIn, tweet, or send an email without your explicit approval. This is a deliberate safety feature, not a limitation.

​
How Draft-Only Works

1

Agent Drafts

Your agent prepares the content — a LinkedIn post, a tweet thread, an email response. It uses your SOUL.md personality and past communication patterns to match your voice.
2

You Review

Drafts appear in your dashboard under the Comms section with a “Pending Approval” status. You can also receive draft previews via Telegram.
3

You Approve or Edit

Approve the draft as-is, edit it before sending, or reject it entirely. Your agent learns from edits to improve future drafts.
4

Agent Sends

Only after your explicit approval does the content get published or sent. The agent logs the action and any edits you made for learning.

​
Why Draft-Only Matters

Founder OS handles channels where a bad message can cause real damage:
  • A poorly worded LinkedIn post reflects on your personal brand
  • A misunderstood email to a prospect can kill a deal
  • A tweet taken out of context can go viral for the wrong reasons
Draft-only mode gives you the productivity benefits of AI-assisted communications without the risk of autonomous posting. Your agent handles the 80% of work (research, drafting, formatting) while you retain the 20% that matters most (judgment, tone, timing).

​
Enabling Auto-Send for Trusted Channels

Auto-send is a power user feature. Only enable it for channels where you trust the output and the consequences of a mistake are low.
If you have calibrated your agent’s voice over time and want to remove the approval step for specific channels, you can enable auto-send in your HUB.yaml: 
integrations:
  telegram:
    mode: "live"             # Already live by default

  linkedin:
    mode: "auto"             # Removes approval step
    guardrails:
      - max_posts_per_day: 3
      - no_controversial_topics: true
      - require_review_if_mentions: ["competitor", "pricing"]

  twitter:
    mode: "draft"            # Keep draft-only for Twitter

  gmail:
    mode: "auto"
    guardrails:
      - auto_send_only: ["follow-up", "scheduling"]
      - draft_required: ["cold-outreach", "pricing", "legal"]
Even in auto-send mode, you can define guardrails that force specific types of content back to draft review. This gives you granular control over what your agent can do autonomously.

​
When to Use Founder OS

Choose Founder OS if you:
  • Spend significant time on GTM — content creation, outreach, pipeline management
  • Need help with LinkedIn and Twitter content at scale
  • Want AI-drafted emails that match your voice
  • Value the safety of draft-only mode for external communications
  • Run an agency and need a GTM operations hub
Choose something else if you:
  • Primarily write code — use Dev OS
  • Want a general-purpose personal OS — use MyOS
  • Do deep research and paper synthesis — use Research OS

​
SOUL.md Personality

Founder OS ships with a GTM-focused agent personality: 
# SOUL.md — Founder OS

You are a GTM-focused AI operating system for {{USERNAME}}.

## Core Traits
- Pipeline-minded — every interaction is a potential opportunity
- Content-native — think in hooks, threads, and narratives
- Safety-first on comms — never send without approval unless auto-send enabled
- Data-driven — track what works, learn from engagement metrics

## Priorities
1. Monitor pipeline and surface deals needing attention
2. Draft content for LinkedIn and Twitter on the content calendar
3. Prepare email responses matching the founder's voice
4. Track engagement metrics and identify what resonates
5. Scan social channels for relevant conversations

## Communication Style
- Telegram for action-required items only
- Dashboard for pipeline overview and content calendar
- Email drafts match the founder's writing style and tone

​
Next Steps

All Templates

Compare all 6 templates side by side

AI OS Overview

How workspaces, integrations, and model routing work