Solanasis — Claude Code + Cowork GTM Master Playbook

Version: 1.0 Date: 2026-03-09 Owner: Dmitri Sunshine, Founder & CEO Goal: Map every Claude Code, Cowork, and Claude in Chrome capability to Solanasis’s go-to-market strategy — marketing, content, outreach, prospecting, and operational leverage.

The Full Tool Landscape (What You Actually Have)
Claude Code — Deep Dive on GTM-Relevant Capabilities
Claude Cowork — Deep Dive on GTM-Relevant Capabilities
Claude in Chrome — Deep Dive on GTM-Relevant Capabilities
Sub-Agents — How to Build Your AI Team
MCP Connectors — Your Integration Layer
Skills & Plugins — Your Capability Library
Scheduled Tasks — Your Automation Engine
Hooks — Your Quality Gates
CLAUDE.md — Your Operating System
The Solanasis GTM Tool Map (Capability → Use Case)
Marketing Playbook — Tool-by-Tool
Outreach & Prospecting Playbook — Tool-by-Tool
Content Factory Workflow — End to End
Competitive Intelligence Engine
Referral Program Automation
Weekly GTM Operating Cadence
Community-Sourced Tips (Reddit, Dev Blogs, Power Users)
Implementation Roadmap (30/60/90 Days)
Tool Cost Summary

1) The Full Tool Landscape

Here’s everything available to you across the three Claude interfaces, mapped to GTM relevance.

The Three Interfaces

Interface	Best For	Access
Claude Code (CLI/Terminal)	Batch automation, scripts, code generation, API integrations, multi-file operations, deploying blog posts, building tools	Terminal / IDE extensions / Desktop app
Claude Cowork (Desktop App)	File creation (docx/pptx/xlsx/pdf), scheduled tasks, plugin-based workflows, non-technical automation, document generation	Claude Desktop App — Cowork mode
Claude in Chrome (Browser Extension)	Live web research, prospect triage, LinkedIn browsing, competitor analysis, form filling, content mining, browser-to-doc handoff	Chrome extension sidebar

Capability Matrix

Capability	Code	Cowork	Chrome
File read/write/edit	Yes	Yes	No
Run bash/scripts	Yes	Yes (sandboxed)	No
Web search	Yes	Yes	Yes (live pages)
MCP connectors (Slack, Gmail, etc.)	Yes	Yes	Limited
Sub-agents (parallel workers)	Yes	Yes	No
Scheduled/recurring tasks	Manual (cron)	Built-in UI	No
Git operations	Yes	No	No
Browser interaction (click, type, navigate)	No	No	Yes
Document generation (docx/pptx/xlsx/pdf)	Via code	Native skills	No
Image generation prompts	Yes	Yes	No
Plugin marketplace	Yes	Yes	No

Pro Tip: The power move is using all three interfaces together. Claude Code builds the automation backbone, Cowork handles document generation and scheduled tasks, and Chrome does the live browser research. Think of it as your AI team: Code is the engineer, Cowork is the operations manager, Chrome is the field researcher.

2) Claude Code — Deep Dive on GTM-Relevant Capabilities

2.1 Built-in Tools (14+ tools out of the box)

These are the core tools Claude Code has access to without any setup:

Read — Read any file on the system
- GTM use: Load prospect lists, voice profiles, brand docs, content seeds
Write — Create new files
- GTM use: Generate blog posts as .md files, create outreach templates, build prospect lists
Edit — Modify existing files with surgical precision
- GTM use: Update CRM exports, refine content drafts, modify website copy
Bash — Run any shell command
- GTM use: Run Python scripts for data enrichment, deploy blog posts via git, process CSVs (Customer/Comma-Separated Values files), automate file operations
Glob — Fast file pattern matching
- GTM use: Find all blog posts, locate prospect files, search content archives
Grep — Content search across files
- GTM use: Search for specific prospects, find content by topic, audit existing materials
WebSearch — Search the web
- GTM use: Competitor research, market research, find prospect info, trending topics
WebFetch — Fetch and process web pages
- GTM use: Pull competitor website copy, extract prospect company info, research industry reports
Agent (Sub-agent spawner) — Launch parallel workers
- GTM use: Run multiple research tasks simultaneously, batch content generation, parallel prospect enrichment
NotebookEdit — Edit Jupyter notebooks
- GTM use: Data analysis dashboards, prospect scoring models, analytics
TodoWrite — Task tracking
- GTM use: Track multi-step GTM workflows
AskUserQuestion — Gather input via structured questions
- GTM use: Qualification questionnaires, content approval workflows

2.2 CLAUDE.md (Your AI Operating System)

CLAUDE.md is the single most important file for your GTM workflow. It’s a markdown file in your project root that Claude Code reads automatically at the start of every session.

What to put in it for GTM:

Your voice profile reference (link to voice-profile.md)
Brand style guide summary
ICP (Ideal Customer Profile) definition
Content pillars
Outreach style rules
CTA standards (go.solanasis.com/meet)
Signature lines (“Backups don’t matter until you restore.“)
File structure conventions
MCP server configurations

Why this matters: Every time you open Claude Code, it already knows your voice, your brand, your ICP, and your rules. No re-explaining. This is the “invest once, harvest forever” Smartcuts approach.

2.3 Sub-Agents (Your AI Team)

Sub-agents are specialized workers that run in parallel, each with their own context window. They’re defined as markdown files with YAML frontmatter (metadata at the top of a file that defines configuration).

GTM-relevant sub-agent types you can create:

Agent Name	Purpose	Tools Access
`prospect-researcher`	Research companies, find fit signals	WebSearch, WebFetch, Read, Write
`content-generator`	Draft blog posts, LinkedIn posts, newsletters	Read, Write, Edit
`seo-optimizer`	Analyze keyword density, meta descriptions, internal links	Read, Edit, Grep, WebSearch
`competitor-analyst`	Monitor competitor websites, extract positioning	WebSearch, WebFetch, Write
`outreach-drafter`	Create personalized outreach messages	Read, Write
`data-enricher`	Process prospect CSVs, add enrichment data	Bash, Read, Write, WebSearch

How they work in practice:

You: "Research these 10 companies and draft personalized outreach for each"

Claude Code (orchestrator):
  → Spawns 10 prospect-researcher sub-agents in parallel
  → Each agent researches one company
  → Results come back to orchestrator
  → Orchestrator spawns outreach-drafter agent
  → Drafter creates personalized messages using research
  → Final output: 10 researched, personalized outreach drafts

Worktree Isolation: Sub-agents can run in their own git worktrees (isolated copies of your repo), meaning they can work simultaneously without stepping on each other’s files. This is especially powerful for batch content generation — you could have 5 agents each writing a different blog post at the same time.

Pro Tip: The key insight from the Claude Code community is that sub-agents prevent “context pollution.” If you ask one agent to research 10 companies in sequence, the context window fills up with company #1’s data by the time it gets to company #10. Sub-agents each get a fresh context window, so company #10 gets the same quality research as company #1.

2.4 MCP (Model Context Protocol) — Integration Layer

MCP is the open standard that lets Claude Code connect to external tools. Think of it as the USB port for AI — any tool that speaks MCP can plug in.

How to set it up:

Configure MCP servers in your Claude Code settings (project-level or global)
Each server exposes “tools” that Claude can call
Tools follow the naming convention: mcp__<server-name>__<tool-name>

Available MCP connectors relevant to Solanasis GTM:

Connector	What It Does for GTM
Slack	Post updates, search channels, read messages
Gmail	Search emails, draft responses, read threads
Google Drive	Search files, create/update docs, organize deliverables
Google Calendar	Check availability, create events for prospect calls
LinkedIn (via Sales Navigator API or browser)	Lead data, connection management
Notion	CRM-style databases, wiki, project management
Asana / ClickUp	Task management, project tracking
Apollo.io	Find leads, enrich lead data, add to sequences
Clay	Lead enrichment, data cleaning, prospect intelligence
Outreach	Sequence management, email automation
Canva	Generate branded graphics, carousels, social media visuals
Figma	Design assets, brand materials
HubSpot / Salesforce	CRM operations, pipeline management
Brevo	Email marketing, newsletter delivery
GitHub	Blog deployment, site management

The Apollo.io connector is a game-changer for Solanasis:

Find leads matching your ICP directly from Claude
Enrich lead data (company size, tech stack, funding)
Add prospects to outreach sequences
All without leaving your Claude workflow

Pro Tip: You can create custom MCP servers for your own tools. If you have a Baserow CRM or a custom lead database, you can build an MCP server that lets Claude read/write directly to it. This turns Claude into a native part of your CRM workflow instead of a separate tool.

2.5 Hooks (Quality Gates & Automation Triggers)

Hooks are Python functions that run at specific points in the Claude agent loop. They’re your quality gates.

GTM-relevant hooks:

Hook Type	When It Fires	GTM Use Case
PreToolUse	Before any tool runs	Validate that outreach drafts include CTA before saving
PostToolUse	After any tool runs	Auto-log all prospect research to a tracking file
PreMessage	Before Claude responds	Inject brand voice rules into every response
PostMessage	After Claude responds	Auto-save all generated content to an archive folder
SessionStart	When a session begins	Load latest prospect pipeline status
SessionEnd	When a session ends	Generate session summary with all work done

Example: Auto-brand-check hook Every time Claude generates content (Write tool), a hook could automatically check it against your voice profile and flag deviations before saving.

2.6 Skills (Reusable Capability Packages)

Skills are instruction sets (SKILL.md files) that Claude follows automatically when relevant. Think of them as SOPs (Standard Operating Procedures) for your AI.

GTM-relevant skills you can build:

content-generator — “When asked to create content, always load voice profile, use content pillars, include CTA, follow publishing cadence”
prospect-researcher — “When researching a company, always check these 8 signals, output in this format, score 1-10”
outreach-drafter — “When drafting outreach, always use these style rules, never use these words, always include one personal observation”
blog-publisher — “When creating a blog post, use this template, include meta description, add to sitemap, deploy via git”
competitive-intel — “When analyzing a competitor, extract these 6 data points, compare against Solanasis positioning”

2.7 Plugins (Packaged Skill + MCP Bundles)

Plugins bundle skills, MCP servers, and tools into installable packages. There are now 50+ in the marketplace.

GTM-relevant plugins available:

Plugin	What It Does
Marketing Skills (coreyhaines31/marketingskills)	42 marketing skills — CRO (Conversion Rate Optimization), copywriting, SEO (Search Engine Optimization), analytics, growth engineering
Sales Plugin (Anthropic official)	Prospecting, outreach, pipeline management, call prep, deal strategy
Legal Plugin (already installed)	Contract review, NDA (Non-Disclosure Agreement) triage, compliance checks — useful for ORB SOWs
Productivity Plugin (already installed)	Task management, memory system — useful for tracking GTM activities
Brand Voice Plugin (already installed)	Brand voice enforcement, guideline generation — critical for content

Pro Tip: The alirezarezvani/claude-skills marketplace has 169 production-ready skills including C-level advisory, compliance, and marketing. You can install the entire marketing suite with one command: /plugin marketplace add alirezarezvani/claude-skills then /plugin install marketing-skills@claude-code-skills. This gives you instant access to skills for SEO audits, conversion optimization, email copywriting, and growth engineering — all of which apply to Solanasis content and outreach.

3) Claude Cowork — Deep Dive on GTM-Relevant Capabilities

3.1 Core Capabilities

Cowork uses the same agentic architecture as Claude Code but runs in the Claude Desktop App with a visual interface. Key differences from Claude Code:

Visual file output — Creates docx, pptx, xlsx, pdf files with professional formatting
Built-in scheduled tasks — Set up recurring automation through the UI
Plugin marketplace — Install capabilities with one click
No terminal required — Accessible for non-technical tasks
Cross-app workflows — Can pass data between Excel and PowerPoint

3.2 Built-in Skills (Document Generation)

Skill	GTM Use Case
xlsx	Create prospect tracking spreadsheets, analytics dashboards, pricing calculators, ORB risk registers
pptx	Create pitch decks, ORB readout presentations, partner program decks, conference slides
docx	Create ORB executive summaries, SOWs, proposals, one-pagers, blog drafts as Word docs
pdf	Create/fill PDF forms, merge ORB deliverables, create partner one-pagers, proposal PDFs
skill-creator	Build custom skills for any repeating Solanasis workflow

3.3 Installed Plugins (Your Current Arsenal)

Based on what’s currently installed in your Cowork environment:

Legal Plugin:

contract-review — Review vendor/client contracts against your playbook
nda-triage — Screen incoming NDAs (Green/Yellow/Red classification)
compliance — GDPR/CCPA navigation, DPA (Data Processing Agreement) reviews
legal-risk-assessment — Classify risks by severity/likelihood
meeting-briefing — Prep structured briefings for client meetings
canned-responses — Template responses for common legal inquiries

Productivity Plugin:

memory-management — Two-tier memory system (CLAUDE.md + memory/ directory) so Claude remembers your context across sessions
task-management — Shared TASKS.md file for tracking commitments

Brand Voice Plugin:

brand-voice-enforcement — Apply your brand guidelines to all content creation
discover-brand — Search connected platforms for brand materials
guideline-generation — Generate brand voice guidelines from source materials

Plugin Management:

create-cowork-plugin — Build new plugins from scratch
cowork-plugin-customizer — Customize existing plugins for Solanasis workflows

3.4 Scheduled Tasks (Your Automation Engine)

This is where Cowork becomes genuinely powerful for GTM. You can create tasks that run automatically on any cadence.

GTM-relevant scheduled tasks to set up:

Task	Cadence	What It Does
Daily Competitive Pulse	Daily (weekdays)	Search web for competitor mentions, pricing changes, new offerings
Weekly Content Batch	Weekly (Monday)	Generate the week’s LinkedIn posts from your content seeds
Weekly Pipeline Review	Weekly (Friday)	Compile prospect status from CRM/spreadsheet into a summary
Bi-Weekly Newsletter Draft	Every 2 weeks	Compile best LinkedIn posts into Substack newsletter draft
Monthly Analytics Report	Monthly (1st)	Pull LinkedIn analytics, blog traffic, lead conversion data into xlsx
Daily LinkedIn Engagement Ideas	Daily	Suggest 5 ICP profiles to comment on based on recent activity
Weekly Referral Check-in	Weekly	Draft follow-up messages for referral partners

How to create a scheduled task:

Type /schedule in Cowork
Describe the task in detail
Choose cadence (daily, weekly, custom cron expression)
Claude creates a SKILL.md file that runs on schedule
Each run has full access to all tools, skills, and connectors

Pro Tip: Scheduled tasks have access to everything — MCP connectors, skills, plugins, the works. This means your “Daily Competitive Pulse” task can search the web, check your connected Slack for relevant mentions, update a Google Sheet, and draft a Slack message with the summary. All automatically. This is the closest thing to having a part-time marketing analyst on staff.

3.5 Sub-Agent Coordination in Cowork

Cowork can spawn sub-agents for complex multi-step tasks. When you ask Cowork to do something complex, it automatically breaks the work into sub-tasks and coordinates parallel workstreams.

Example GTM workflow:

You: "Create a complete outreach package for [Company X]"

Cowork:
  → Sub-agent 1: Research company website and LinkedIn
  → Sub-agent 2: Check for news/press about the company
  → Sub-agent 3: Analyze their tech stack and operational maturity
  → Orchestrator: Synthesize findings into account brief
  → Sub-agent 4: Draft 3 personalized outreach messages
  → Sub-agent 5: Create a one-page company-specific proposal overview
  → Final output: Complete outreach package ready for your review

4) Claude in Chrome — Deep Dive on GTM-Relevant Capabilities

4.1 What It Can Do

Claude in Chrome is a browser copilot that can see what’s on your screen, navigate pages, fill forms, extract data, and help you work through browser-based tasks.

Core capabilities:

Read any visible page — Understand page content, extract structured data
Navigate — Click links, scroll, switch tabs
Multi-tab synthesis — Compare information across multiple open tabs
Form interaction — Fill forms, select dropdowns (with permission guardrails)
Screenshot analysis — See and reason about visual content
Structured extraction — Turn messy browser content into clean data
Workflow recording — Save repeatable browser workflows as shortcuts

4.2 GTM-Specific Browser Workflows

These are your highest-leverage Claude in Chrome workflows:

LinkedIn Prospect Triage
- Open Sales Navigator search results
- Claude reads visible profiles, scores fit against ICP
- Outputs ranked shortlist with fit reasons and next actions
Account Brief Creation
- Open company website + LinkedIn + any other relevant tabs
- Claude synthesizes into structured account brief
- Includes: what they do, likely pains, operational maturity signals, best outreach angle
Competitor Website Analysis
- Open 3-5 competitor sites in tabs
- Claude extracts positioning, pricing, target market, messaging gaps
- Outputs comparison matrix + Solanasis differentiation points
Content Mining from LinkedIn Feed
- Browse your LinkedIn feed or specific hashtags
- Claude identifies recurring pain points, language patterns, trending topics
- Outputs content ideas mapped to your content pillars
Browser-to-CRM Handoff
- After researching a prospect across multiple tabs
- Claude structures everything into CRM-ready format
- Copy/paste into your tracking system

4.3 Shortcuts & Recorded Workflows

Claude in Chrome supports saved shortcuts — reusable workflows you can trigger with a command. You can list available shortcuts and create new ones.

Recommended shortcuts to build:

/prospect-triage — Run ICP scoring on current page
/account-brief — Generate structured account brief from current tabs
/outreach-draft — Draft personalized outreach from current context
/competitor-scan — Extract competitor positioning from current page
/content-mine — Pull content ideas from current feed/page

4.4 GIF Recording (Demo Creation)

Chrome can record your browser interactions as GIF files — useful for:

Creating demo workflows for contractor training
Documenting SOPs visually
Creating social media content showing “behind the scenes”
Building training materials for referral partners

5) Sub-Agents — How to Build Your AI Team

5.1 Agent Types for Solanasis GTM

You can define custom sub-agents as markdown files. Here’s the recommended Solanasis AI team:

Agent: Explore (Built-in)

Purpose: Fast codebase/file exploration
GTM use: Quickly find content, templates, prospect data across your workspace
Thoroughness levels: quick, medium, very thorough

Agent: Plan (Built-in)

Purpose: Design implementation strategies
GTM use: Plan content campaigns, outreach sequences, website updates

Agent: General-Purpose (Built-in)

Purpose: Complex multi-step research and execution
GTM use: Deep prospect research, multi-source competitive analysis

Custom Agent: prospect-enricher

---
description: "Enrich prospect data with web research"
tools: [WebSearch, WebFetch, Read, Write]
model: claude-sonnet-4-6
---
Research the given company and contact. Find: company size, industry, tech stack clues, recent news, operational maturity signals, security posture indicators. Output as structured markdown.

Custom Agent: content-writer

---
description: "Generate brand-voice content"
tools: [Read, Write, Edit]
model: claude-opus-4-6
---
Load the voice profile from solanasis-docs/solanasis-voice-profile.md and brand style guide. Generate content following the content pillars and style rules. Always include a soft CTA.

5.2 Parallel Execution Patterns

Pattern 1: Fan-out Research

Spawn 5-10 research agents simultaneously
Each researches one prospect
Orchestrator collects and ranks results

Pattern 2: Content Factory

Spawn agents for LinkedIn post, blog post, tweet thread, newsletter simultaneously
Each works from the same content seed
Orchestrator reviews and packages all content

Pattern 3: Multi-Source Intelligence

Agent 1 researches competitor websites
Agent 2 searches for competitor mentions on social media
Agent 3 checks industry reports and news
Orchestrator synthesizes into competitive intel briefing

Pro Tip: Boris Cherny (Anthropic engineer) announced in February 2026 that sub-agents now support worktree isolation. This means each sub-agent can work in its own isolated copy of your files. For Solanasis, this is huge — you could have 5 agents each writing a different blog post simultaneously, each in their own worktree, and then merge them all together. No conflicts, no overwriting.

6) MCP Connectors — Your Integration Layer

6.1 Available Connectors (50+ as of March 2026)

Communication & Collaboration:

Slack, Gmail, Microsoft 365/Outlook, Google Chat

File Storage & Docs:

Google Drive, Notion, Box, SharePoint, Confluence

Project Management:

Asana, ClickUp, Linear, Jira, Monday.com

CRM & Sales:

HubSpot, Salesforce, Apollo.io, Clay, Outreach, Pipedrive

Design:

Canva, Figma

Engineering:

GitHub, GitLab, Vercel

Analytics:

Google Analytics (via API), various data connectors

Custom:

Any MCP-compatible server you build or find

6.2 Priority Connectors for Solanasis GTM

Tier 1 (Set Up Immediately):

Google Drive — Store/retrieve ORB deliverables, proposals, one-pagers
Gmail — Draft/read outreach emails, follow-ups, partner communications
Slack (if using) — Team communications, automated notifications
GitHub — Blog deployment, website updates
Google Calendar — Scheduling awareness for prospect calls

Tier 2 (Set Up in 30 Days): 6. Apollo.io — Lead finding, enrichment, sequence management 7. Canva — Branded graphic generation 8. ClickUp — Task tracking integration

Tier 3 (Set Up in 60 Days): 9. Brevo — Newsletter automation 10. Clay — Advanced lead enrichment 11. LinkedIn (unofficial/browser-based) — Via Chrome for research

6.3 The Apollo.io Power Play

Apollo.io announced a direct Claude integration in 2026. This is particularly relevant for Solanasis because:

Find leads matching your ICP criteria directly from Claude
Enrich data — Get company size, tech stack, funding info
Add to sequences — Put prospects into outreach sequences without switching tools
Search by signals — Find companies with specific triggers (new funding, leadership changes, compliance events)

Example workflow:

"Find 20 Colorado SMBs with 11-50 employees in professional services
that recently changed their IT leadership. Enrich with contact info
for the CEO and COO. Draft personalized outreach for each using our
voice profile and the restore-test hook."

7) Skills & Plugins — Your Capability Library

7.1 Skills You Already Have

From your current Cowork environment:

Skill	Category	GTM Relevance
xlsx	Document	Prospect tracking, analytics, pricing calculators
pptx	Document	Pitch decks, ORB readouts, partner decks
docx	Document	Proposals, SOWs, exec summaries, one-pagers
pdf	Document	ORB deliverables, partner one-pagers, form filling
skill-creator	Meta	Build custom skills for any workflow
schedule	Automation	Create recurring automated tasks
brand-voice-enforcement	Content	Apply your voice to all content
discover-brand	Content	Find brand materials across platforms
guideline-generation	Content	Generate/refine brand voice guidelines
contract-review	Legal	Review ORB agreements, vendor contracts
nda-triage	Legal	Screen partner/client NDAs
compliance	Legal	Privacy regulation navigation
legal-risk-assessment	Legal	Risk scoring for deals
meeting-briefing	Legal/Sales	Prep for client/prospect meetings
memory-management	Productivity	Persistent context across sessions
task-management	Productivity	Track commitments and next actions
create-cowork-plugin	Meta	Build new plugins
cowork-plugin-customizer	Meta	Customize plugins for Solanasis

7.2 Skills You Should Build

Skill Name	What It Does	Priority
`orb-proposal-generator`	Generate complete ORB proposals from prospect data	High
`prospect-scorecard`	Score prospects against ICP with weighted criteria	High
`linkedin-post-generator`	Generate LinkedIn posts from content seeds using voice profile	High
`blog-post-factory`	Create full blog posts with SEO optimization for your Astro site	High
`outreach-sequence-builder`	Create multi-touch outreach sequences for different prospect types	Medium
`competitive-briefing`	Generate structured competitor analysis reports	Medium
`referral-partner-kit`	Generate personalized partner materials from templates	Medium
`weekly-gtm-report`	Compile weekly GTM metrics into a formatted report	Medium
`orb-deliverable-builder`	Generate ORB exec summaries, risk registers, and 30/60/90 plans	High
`contractor-sop-generator`	Create step-by-step SOPs for contractor-delivered work	Low

7.3 Marketplace Plugins Worth Installing

From the Official Anthropic Plugin Store:

Sales Plugin — Prospecting, outreach, pipeline management, call prep
Marketing Plugin — Campaign planning, content strategy, analytics

From Community Marketplaces:

marketingskills (coreyhaines31) — CRO, copywriting, SEO, analytics, growth engineering
claude-skills (alirezarezvani) — 169 skills including C-level advisory, marketing, compliance

Pro Tip: You can create your own plugin that bundles all your Solanasis-specific skills, MCP configurations, and brand docs into a single installable package. This means if you ever set up a new machine, add a contractor, or want to share your workflow with a partner, it’s one install. Use the create-cowork-plugin skill to build it.

8) Scheduled Tasks — Your Automation Engine

8.1 How Scheduled Tasks Work

Created via /schedule in Cowork or programmatically
Run on a cadence: daily, weekly, custom cron, or on-demand
Full access to all tools, skills, connectors, and plugins
Each run starts a fresh session but follows the task prompt
Output can be saved to files, sent via connectors, or both

8.2 Recommended Solanasis GTM Scheduled Tasks

DAILY Tasks:

Task ID	Cadence	Description	What It Does
`daily-competitive-pulse`	Weekdays 8:00 AM	Competitor monitoring	Web search for competitor mentions, pricing changes, new content. Save to competitive-intel/ folder.
`daily-content-ideas`	Weekdays 7:30 AM	Content inspiration	Search trending cybersecurity, SMB ops, and DR topics. Output 3 content angles with hooks.
`daily-linkedin-engagement`	Weekdays 9:00 AM	Engagement suggestions	Identify 5 ICP-relevant LinkedIn profiles/posts to engage with. Output talking points.

WEEKLY Tasks:

Task ID	Cadence	Description	What It Does
`weekly-content-batch`	Mondays 6:00 AM	Content generation	Load content seeds + voice profile. Generate 4-5 LinkedIn posts + 1 blog draft for the week.
`weekly-pipeline-review`	Fridays 4:00 PM	Pipeline summary	Compile prospect status, recent touchpoints, next actions into a summary report.
`weekly-referral-nudge`	Wednesdays 10:00 AM	Referral follow-up	Draft check-in messages for referral partners. Personalize based on recent activity.
`weekly-seo-check`	Saturdays 8:00 AM	SEO monitoring	Check blog post rankings, suggest optimizations, identify new keyword opportunities.

BI-WEEKLY Tasks:

Task ID	Cadence	Description	What It Does
`biweekly-newsletter`	Every other Friday	Newsletter draft	Compile best LinkedIn content into Substack newsletter draft with personal commentary prompts.

MONTHLY Tasks:

Task ID	Cadence	Description	What It Does
`monthly-analytics`	1st of month	Analytics report	Pull all GTM metrics (LinkedIn, blog, email, pipeline) into formatted xlsx report.
`monthly-competitive-report`	1st of month	Competitive analysis	Compile daily competitive intel into monthly trends report with strategic implications.
`monthly-content-calendar`	Last day of month	Next month planning	Generate next month’s content calendar with topics, hooks, and draft outlines.

8.3 Cron Expression Cheat Sheet

For custom scheduling in Cowork:

Schedule	Cron Expression
Every weekday at 9 AM	`0 9 * * 1-5`
Every Monday at 6 AM	`0 6 * * 1`
Every Friday at 4 PM	`0 16 * * 5`
1st of every month at 8 AM	`0 8 1 * *`
Every 2 hours during business hours	`0 9-17/2 * * 1-5`

9) Hooks — Your Quality Gates

9.1 GTM-Relevant Hook Patterns

Content Quality Gate:

Trigger: Before saving any content file (PreToolUse on Write/Edit)
Action: Check content against voice profile rules
Block if: Contains banned phrases, missing CTA, too long/short, sounds generic

Outreach Safety Gate:

Trigger: Before sending any outreach (PreToolUse)
Action: Verify personalization, check for template language, verify CTA
Block if: Sounds like mass blast, missing personal observation, too long

Data Logging Hook:

Trigger: After any prospect research (PostToolUse)
Action: Auto-append research output to prospect-log.md with timestamp

Session Summary Hook:

Trigger: End of session (SessionEnd)
Action: Generate summary of all GTM work done, next actions, files created

9.2 Implementing Hooks

Hooks are defined in your Claude Code settings as Python functions. Here’s a conceptual example:

# Pre-write hook: Ensure all content includes CTA
def pre_write_check(tool_call):
    if "content" in tool_call.path:
        if "solanasis.com/meet" not in tool_call.content:
            return "WARNING: Content missing CTA. Add booking link."

Pro Tip: Start with logging hooks before enforcement hooks. Log everything for the first month so you can see patterns. Then build enforcement hooks based on the patterns you see (most common mistakes, most frequent deviations from brand voice, etc.). This is the “observe, then automate” approach that prevents over-engineering.

10) CLAUDE.md — Your Operating System

10.1 What Goes in CLAUDE.md

Your CLAUDE.md should be the master configuration file that every Claude session loads. Here’s the recommended structure for Solanasis GTM:

# Solanasis Project Instructions
 
## Identity
- Company: Solanasis LLC (fCIO/fCSIO/fCOO for SMBs and nonprofits)
- Founder: Dmitri Sunshine
- Tagline: "Operational Resilience, Proven"
- Website: solanasis.com
 
## Voice & Style
- Load voice profile from: solanasis-docs/solanasis-voice-profile.md
- Primary voice: Bold Contrarian (founder-led)
- Secondary voice: Executive Calm (board-level content)
- Never: sensationalized tones, jargon without explanation, more than 3 sentences per paragraph
- Signature lines: [list your signature lines]
 
## ICP
- Primary: 10-150 seat SMBs + nonprofits on M365/Google Workspace
- Buyer: CEO/Executive Director
- Champion: Ops lead + IT/MSP contact
- Geography: Colorado (expanding to West Coast)
 
## Content Rules
- Every piece maps to a business outcome
- Always include soft CTA (go.solanasis.com/meet)
- Content pillars: [list pillars]
- Use bullet points and numbered lists for readability
 
## Outreach Rules
- Warm, sharp, grounded, not salesy
- Short paragraphs
- Include one personal/specific observation
- Never sound like an SDR or template
 
## File Structure
- Content: solanasis-docs/content-creation/
- Prospects: solanasis-docs/prospects/
- Deliverables: solanasis-docs/deliverables/
- Blog: solanasis-site/src/content/blog/

10.2 Multiple CLAUDE.md Files

You can have different CLAUDE.md files in different directories:

Root-level: Global rules for all Solanasis work
solanasis-site/CLAUDE.md: Website-specific instructions (already exists)
solanasis-docs/content-creation/CLAUDE.md: Content-specific rules
solanasis-docs/prospects/CLAUDE.md: Prospecting-specific rules

11) The Solanasis GTM Tool Map (Capability → Use Case)

Marketing Use Cases

Use Case	Best Tool	Why
Generate LinkedIn posts	Cowork + brand-voice skill	Voice enforcement + easy review
Generate blog posts	Claude Code + blog-publisher skill	Direct git commit to deploy
Create pitch decks	Cowork + pptx skill	Professional PowerPoint output
Create one-pagers	Cowork + pdf skill	Clean PDF output
Design social graphics	Cowork + Canva MCP	AI-driven branded design
Write newsletter	Cowork + scheduled task	Automated bi-weekly compilation
SEO optimization	Claude Code + WebSearch	Technical analysis + implementation
Content calendar planning	Cowork + scheduled task	Monthly automated planning
Image prompt generation	Claude Code or Cowork	Generate prompts for DALL-E/Midjourney
Brand voice enforcement	Cowork + brand-voice-enforcement skill	Ensures all content sounds like you

Outreach & Prospecting Use Cases

Use Case	Best Tool	Why
Prospect research	Chrome + Claude Code sub-agents	Live page reading + parallel deep research
Account brief creation	Chrome + Cowork	Browser context → structured document
Outreach message drafting	Claude Code + outreach-drafter agent	Batch personalized messaging
Lead enrichment	Claude Code + Apollo.io MCP	Direct CRM integration
CRM updates	Claude Code + MCP connector	Automated data entry
Competitor monitoring	Cowork scheduled task	Daily automated intelligence
Referral partner outreach	Cowork + brand-voice skill	On-brand partner communications
Sales Navigator list building	Chrome	Live browser interaction
Pipeline tracking	Cowork + xlsx skill	Formatted spreadsheet reports
Pre-call research	Chrome + meeting-briefing skill	Structured briefing document

Operational / Delivery Use Cases

Use Case	Best Tool	Why
ORB proposal generation	Cowork + docx/pdf skills	Professional document output
ORB executive summary	Cowork + pdf skill	Board-ready PDF
Risk register creation	Cowork + xlsx skill	Formatted Excel with formulas
30/60/90 action plan	Cowork + xlsx skill	Formatted Excel
Contract review	Cowork + contract-review skill	Legal plugin analysis
NDA screening	Cowork + nda-triage skill	Green/Yellow/Red classification
Meeting prep	Cowork + meeting-briefing skill	Structured briefing
Contractor SOPs	Claude Code + Write	Markdown SOPs in workspace

12) Marketing Playbook — Tool-by-Tool

12.1 Content Creation Flow (End-to-End)

Step 1: Seed Generation

Tool: Cowork (conversation) or Claude Code
Input: Your raw ideas, hot takes, client observations
Output: content-seeds-YYYY-MM.md

Step 2: Content Batch Generation

Tool: Claude Code with sub-agents + brand-voice skill
Process:
1. Load CLAUDE.md (auto)
2. Load voice profile + content pillars
3. Spawn sub-agents in parallel:
  - Agent 1: 4-5 LinkedIn posts
  - Agent 2: 1 blog post (long-form)
  - Agent 3: Tweet thread versions
  - Agent 4: 2 carousel outlines
  - Agent 5: Newsletter draft
4. Orchestrator reviews output for consistency
5. Save to content-batch-YYYY-MM/ folder

Step 3: Review & Refinement

Tool: Cowork (MD artifact review)
Process: You review, add personal anecdotes, approve/revise
Time: 45-60 minutes

Step 4: Visual Asset Creation

Tool: Cowork + Canva MCP connector
Process: Generate carousel graphics, blog hero images from templates
Time: 15 minutes (AI does the design)

Step 5: Blog Deployment

Tool: Claude Code + Git
Process: Commit blog .md file → auto-deploy via Cloudflare Pages
Time: 0 minutes (automated)

Step 6: Scheduling

Tool: Claude in Chrome + scheduling tool (Typefully/Buffer)
Process: Load approved content into scheduler
Time: 15 minutes

12.2 SEO Engine

Using Claude Code for SEO:

Keyword research: WebSearch + WebFetch to identify target keywords for each content pillar
On-page optimization: Read blog post, analyze keyword density, suggest improvements via Edit
Meta description generation: Generate SEO-optimized meta descriptions for every blog post
Internal linking: Grep across all blog posts, suggest internal link opportunities
Competitor SEO analysis: WebFetch competitor blog posts, extract keyword patterns
Schema markup: Generate structured data for blog posts (Article, FAQService, etc.)

Scheduled SEO task (weekly):

Check Google Search Console data (via API or manual export)
Identify top-performing and underperforming posts
Suggest optimization actions
Output: weekly SEO report

LinkedIn (primary channel):

Content generation: Claude Code sub-agents
Voice enforcement: brand-voice-enforcement skill
Engagement: Claude in Chrome for comment suggestions
Analytics: Scheduled task pulling metrics weekly
Scheduling: Chrome for Typefully/Buffer, or native LinkedIn scheduler

Substack Newsletter:

Draft generation: Scheduled bi-weekly task in Cowork
Compilation: Auto-select best LinkedIn content from past 2 weeks
Personalization: You add personal commentary sections
Publishing: Manual (or Chrome for Substack scheduling)

X/Twitter:

Repurpose: Claude Code converts LinkedIn posts to tweet threads
Scheduling: Chrome for Typefully

Instagram:

Carousels: Cowork + Canva MCP for branded graphics
Captions: Claude Code generates from LinkedIn content
Scheduling: Chrome for Later/Buffer

13) Outreach & Prospecting Playbook — Tool-by-Tool

13.1 The Full Prospecting Pipeline

DISCOVERY (Chrome + Claude Code)
  │
  ├── Sales Navigator search → Chrome reads results
  ├── Directory/Google search → Chrome reads results
  ├── Apollo.io search → MCP connector returns leads
  │
  ▼
ENRICHMENT (Claude Code sub-agents)
  │
  ├── Company research agent → Website + news analysis
  ├── Contact research agent → LinkedIn profile + recent activity
  ├── Tech stack agent → Identify tools/platforms used
  │
  ▼
QUALIFICATION (Claude Code + scoring skill)
  │
  ├── Score against ICP criteria (1-10)
  ├── Identify trigger events (leadership change, growth, incident)
  ├── Flag fit signals (operational complexity, systems sprawl)
  │
  ▼
MESSAGE PREP (Claude Code + outreach-drafter agent)
  │
  ├── 3 connection note options
  ├── 2 follow-up messages
  ├── 1 warm referral ask
  ├── 1 "saw this and thought of you" note
  │
  ▼
HUMAN REVIEW & SEND (You)
  │
  ├── Review drafts
  ├── Personalize with your observations
  ├── Send via LinkedIn / Email
  │
  ▼
LOGGING (Claude Code + MCP)
  │
  └── Auto-log to CRM/spreadsheet with all context

13.2 Prospect Research Deep Dive

What Claude can research for each prospect:

Company fundamentals: Size, industry, location, revenue clues, growth stage
Operational maturity signals: Website quality, tech stack clues, job postings (hiring IT = growing pains)
Security posture indicators: Privacy policy quality, SSL (Secure Sockets Layer — the lock icon in browsers) implementation, cookie handling
Pain indicators: Outdated website, multiple disconnected tools listed, recent leadership changes
Trigger events: New funding, acquisitions, regulatory changes, security incidents
Decision maker mapping: CEO/ED, COO/Ops Director, IT lead, office manager

Using Apollo.io MCP for enrichment:

Company firmographics (size, revenue, industry)
Contact direct emails and phones
Technology stack data
Funding history
Job openings (signals growth/pain)
Intent data (if available)

13.3 Outreach Message Templates (AI-Enhanced)

The AI doesn’t write the template — it personalizes the template based on research:

Connection Request (Restore Hook):

Input to Claude: "Personalize this for [Company] based on the account brief"

Template: "Quick question [FirstName]: when was [CompanyName]'s last
*real* restore test? [PersonalizedObservation based on research]"

Example output: "Quick question Sarah: when was Bright Future Nonprofit's
last real restore test? I noticed you recently migrated to Microsoft 365 —
that's usually when restore assumptions break."

Follow-Up After Accept:

Input to Claude: "Draft follow-up based on their profile and our brief"

Claude generates: "Thanks for connecting, Sarah. Reason I asked about
restores — we see a lot of nonprofits your size discover gaps right after
a platform migration. We run a 10-day Resilience Checkup that includes
a real restore test + 30/60/90 plan. Want the one-pager?"

13.4 MSP Partnership Outreach

Using Claude to build MSP partner pipeline:

Research phase (Chrome): Browse MSP directories, LinkedIn MSP groups
Enrichment phase (Claude Code): Research each MSP’s client base, service offerings, gaps
Qualification (Claude Code): Score MSPs on likelihood of partnership fit
Message prep (Claude Code): Personalized outreach that positions ORB as non-competitive value-add
Track (Cowork): Pipeline spreadsheet with partner status, follow-up dates

14) Content Factory Workflow — End to End

14.1 Monthly Content Factory (2-3 Hours Total)

Phase 1: Seed (15 min — You)

Jot raw ideas into content-seeds-YYYY-MM.md
Voice memos transcribed by Claude
Hot takes from client interactions (anonymized)
Industry observations

Phase 2: Generate (0 min — Automated)

Scheduled task or manual Claude Code session
Loads: voice profile + brand guide + content pillars + seeds
Spawns parallel sub-agents:
- LinkedIn posts (4-5 per week)
- Blog post (1 per week)
- Tweet threads (from LinkedIn posts)
- Carousel outlines (2-3 per month)
- Newsletter draft (bi-weekly)

Phase 3: Review (45-60 min — You)

Read all content in Cowork MD artifacts
Add personal anecdotes and real observations
Flag revisions, approve the rest

Phase 4: Visual (15 min — Canva MCP)

Generate carousel graphics
Generate blog hero images
Apply brand palette (Ink Navy + Parchment + Copper)

Phase 5: Deploy (15 min — Automated)

Blog: Git commit → Cloudflare auto-deploy
LinkedIn: Load into Typefully/Buffer
Newsletter: Load into Substack
Instagram: Load into Later

Phase 6: Engage (15 min/day — You)

Reply to comments
Comment on 3-5 ICP posts
This is the irreplaceable human element

14.2 Content Repurposing Chain

1 Blog Post (1000+ words)
  ├── → 3-4 LinkedIn Posts (each covering one section)
  ├── → 1 Tweet Thread (key points)
  ├── → 1 Carousel (5-8 slides)
  ├── → 1 Newsletter Section
  ├── → 2-3 Pull Quotes (for Instagram/social)
  └── → 1 Email Snippet (for outreach relevance)

Claude Code can automate this entire repurposing chain with a single prompt + sub-agents.

15) Competitive Intelligence Engine

15.1 Automated Competitive Monitoring

Daily (Scheduled Task):

Web search for competitor brand mentions
Check competitor websites for changes (new services, pricing, blog posts)
Monitor industry keywords for trending discussions
Save findings to competitive-intel/daily/YYYY-MM-DD.md

Weekly (Scheduled Task):

Synthesize daily findings into weekly competitive brief
Identify positioning opportunities
Flag threats or pricing changes
Output: competitive-intel/weekly/YYYY-WXX.md

Monthly (Scheduled Task):

Comprehensive competitor positioning analysis
Market trend synthesis
Solanasis messaging refinement recommendations
Output: competitive-intel/monthly/YYYY-MM.md

15.2 Competitor Categories to Monitor

Based on your ORB Playbook, monitor these categories:

Fractional CIO/CISO firms — Direct competitors
Managed Service Providers (MSPs) — Potential partners AND competitors
Cybersecurity consultancies — Adjacent, may overlap on security assessments
DR/BCP (Business Continuity Planning) providers — Overlap on backup/restore
AI implementation consultants — Overlap on Responsible AI
Nonprofit tech consultants — ICP overlap

15.3 What to Extract

For each competitor, Claude should track:

What they lead with (headline, hero section)
How they price/package
Who they target (ICP clues)
Their messaging language
What feels generic/stale
Where Solanasis can differentiate
Any new offerings, blog posts, case studies

16) Referral Program Automation

16.1 Automated Referral Workflows

Partner Onboarding:

You identify a potential partner
Claude drafts personalized Founding Partner invite (using referral playbook templates)
You review and send
On acceptance: Claude generates partner kit (one-pager, terms, deal registration form)
Auto-add to partner tracking spreadsheet

Deal Registration Processing:

Partner submits deal registration
Scheduled task checks for new registrations daily
Claude validates against qualified referral criteria
Generates acknowledgment email draft
Calculates protection window (today + 120 days)
Updates tracking spreadsheet
Creates prospect record with partner attribution

Commission Tracking:

Scheduled monthly task
Checks all referred deals against payment status
Calculates commission amounts (10% standard / 15% founding)
Flags conversion bonus eligibility ($500 for recurring within 60 days)
Generates quarterly payout report

16.2 Partner Communication Automation

Scheduled bi-weekly task:

Review all active partner deals
Draft personalized status update for each partner
Include: deal stage, expected timeline, any blockers
You review and send

17) Weekly GTM Operating Cadence

Monday — Content & Planning

Time	Activity	Tool
7:30 AM	Daily content ideas (auto)	Scheduled task
8:00 AM	Review content batch for the week	Cowork MD artifacts
8:30 AM	Schedule LinkedIn posts for the week	Chrome + Typefully
9:00 AM	15-min LinkedIn engagement	Chrome

Tuesday — Outreach

Time	Activity	Tool
8:00 AM	Competitive pulse (auto)	Scheduled task
8:30 AM	Prospect research session (10 new leads)	Chrome + Claude Code
9:30 AM	Draft outreach messages	Claude Code
10:00 AM	Review and send outreach	Manual
10:30 AM	15-min LinkedIn engagement	Chrome

Wednesday — Partnerships & Follow-ups

Time	Activity	Tool
8:00 AM	Competitive pulse (auto)	Scheduled task
8:30 AM	Follow up on pending outreach	Chrome + Claude Code
9:00 AM	Referral partner check-ins (auto draft)	Scheduled task
9:30 AM	Review and send partner messages	Manual
10:00 AM	MSP partnership research	Chrome
10:30 AM	15-min LinkedIn engagement	Chrome

Thursday — Deep Work & Content

Time	Activity	Tool
8:00 AM	Competitive pulse (auto)	Scheduled task
8:30 AM	Blog post review and deployment	Claude Code + Git
9:00 AM	SEO optimization on recent posts	Claude Code
9:30 AM	Pre-call research for upcoming meetings	Chrome + meeting-briefing skill
10:30 AM	15-min LinkedIn engagement	Chrome

Friday — Analysis & Prep

Time	Activity	Tool
8:00 AM	Competitive pulse (auto)	Scheduled task
8:30 AM	Weekly pipeline review (auto)	Scheduled task
9:00 AM	Review pipeline report, update statuses	Cowork
9:30 AM	Plan next week’s content seeds	Cowork conversation
10:00 AM	Generate ORB deliverables for active clients	Cowork + skills
10:30 AM	15-min LinkedIn engagement	Chrome

18) Community-Sourced Tips (Reddit, Dev Blogs, Power Users)

18.1 From the Claude Code GTM Community

Tip 1: “Build context that compounds”

Source: Databar.ai Claude Code GTM Guide
Insight: Every research session should save its output to a structured file. Over time, your research directory becomes a proprietary intelligence database that makes every subsequent session smarter.
Solanasis application: Save all prospect research to solanasis-docs/prospects/[company-name].md. Claude loads these automatically in future sessions.

Tip 2: “Mirror the prospect’s language”

Source: MarketBetter.ai LinkedIn outreach guide
Insight: Claude excels at adopting language from source material. Feed it the prospect’s own LinkedIn posts and company copy, and it will naturally mirror their vocabulary in outreach.
Solanasis application: Include prospect’s recent LinkedIn posts in the outreach-drafter agent’s context.

Tip 3: “Teams report 300x output increase for complex tasks”

Source: Firecrawl/Claude Code marketing guide
Insight: The real leverage isn’t in one-off tasks — it’s in building reusable workflows that compound. One good content workflow replaces a $5-10k/month content team.
Solanasis application: Your content factory workflow (Section 14) is designed for this exact multiplier.

Tip 4: “A team that builds its own tools moves faster”

Source: Claude Code for business guides
Insight: Custom MCP servers, skills, and plugins give you proprietary advantages that competitors using off-the-shelf tools can’t replicate.
Solanasis application: Build a Solanasis-specific plugin (using create-cowork-plugin) that bundles your voice profile, ICP criteria, outreach templates, and scoring rubrics.

Tip 5: “Scheduled tasks are the closest thing to having a part-time analyst”

Source: aiblewmymind.substack.com — 6 ways automated work with scheduled tasks
Insight: The power isn’t in individual tasks — it’s in the compounding effect of daily/weekly automated intelligence gathering.
Solanasis application: Your daily competitive pulse + weekly content batch + monthly analytics = a self-running GTM intelligence engine.

Tip 6: “Use the memory management plugin to build persistent context”

Source: Power user reports on Cowork productivity plugins
Insight: The memory-management plugin (CLAUDE.md + memory/ directory) means Claude acts like a colleague who remembers your shorthand, acronyms, and inside references.
Solanasis application: Teach Claude your abbreviations (ORB = Operational Resilience Baseline, DR = Disaster Recovery, ICP = Ideal Customer Profile) and it will use them consistently.

18.2 From Anthropic Engineering Posts

Tip 7: “Sub-agents prevent context pollution”

Source: Boris Cherny (Anthropic), February 2026
Insight: When researching multiple prospects in one session, use sub-agents so each prospect gets a fresh context window. This prevents quality degradation on later prospects.

Tip 8: “Worktree isolation for batch content”

Source: Boris Cherny announcement, February 2026
Insight: Sub-agents + worktrees = parallel content generation without conflicts. 5 blog posts written simultaneously, each in isolation.

Tip 9: “Skills are better than prompts for repeating workflows”

Source: Claude Code Skills documentation
Insight: A SKILL.md file is loaded automatically when relevant, unlike prompts which you have to remember to include. Skills are “always-on expertise.”

18.3 From the Marketing Community

Tip 10: “The Canva MCP connector eliminates the design bottleneck”

Source: Multiple Cowork guides
Insight: You can say “create a LinkedIn carousel about the 5 signs you need DR verification using Solanasis brand colors” and Claude generates it in Canva.

Tip 11: “Apollo + Claude = prospecting on autopilot”

Source: Apollo.io press release, February 2026
Insight: Apollo’s MCP connector lets Claude find leads, enrich data, and add to sequences without you ever opening Apollo’s UI.

Tip 12: “Build a ‘content seed’ ritual — 15 minutes changes everything”

Source: Multiple content strategy blogs
Insight: The bottleneck in AI content isn’t generation — it’s having authentic raw material. 15 minutes of jotting hot takes and observations gives Claude enough to generate a month of content.

19) Implementation Roadmap (30/60/90 Days)

Days 1-30: Foundation

Week 1: Core Setup

Complete voice profile if not finished (Priority #1)
Configure CLAUDE.md with all GTM rules
Install Sales and Marketing plugins
Set up memory-management plugin with Solanasis terminology
Create prospect-scorecard skill
Create linkedin-post-generator skill

Week 2: Content Engine

Run first content batch generation session
Deploy first Claude-generated blog post
Set up weekly-content-batch scheduled task
Set up daily-content-ideas scheduled task
Create 3-5 Canva templates for recurring content types
Connect Canva MCP if using Canva Pro

Week 3: Prospecting Engine

Build prospect-enricher sub-agent
Build outreach-drafter sub-agent
Create Chrome shortcuts for prospect-triage and account-brief
Run first Chrome-based prospect research session (10 prospects)
Set up prospect tracking spreadsheet or CRM fields

Week 4: Automation & Iteration

Set up daily-competitive-pulse scheduled task
Set up weekly-pipeline-review scheduled task
Create first batch of personalized outreach messages
Review first month’s content performance
Refine voice profile based on what sounds right

Days 31-60: Scale

Week 5-6: Expand Prospecting

Connect Apollo.io MCP connector
Build MSP partnership research workflow
Create referral-partner-kit skill
Send first 10 Founding Partner invitations
Set up weekly-referral-nudge scheduled task
Scale to 20+ prospect research per week

Week 7-8: Content Maturation

Launch Substack newsletter with bi-weekly cadence
Set up biweekly-newsletter scheduled task
Create blog-post-factory skill with SEO optimization
Start carousel content on LinkedIn
Begin X/Twitter repurposing
Set up weekly-seo-check scheduled task

Days 61-90: Optimize

Week 9-10: Intelligence & Optimization

Set up monthly-analytics scheduled task
Set up monthly-competitive-report scheduled task
Build competitive intelligence dashboard (xlsx)
First quarterly content review (what’s working?)
Refine ICP scoring criteria based on real data

Week 11-12: Systemize

Create Solanasis master plugin (bundles all skills + config)
Document all SOPs for contractor handoff
Set up contractor-specific CLAUDE.md configurations
Review and refine all scheduled tasks
Plan Month 4+ expansion (new geographies, verticals)

20) Tool Cost Summary

Tool	Monthly Cost	Category	Priority
Claude Max Plan	~$200	AI engine (includes Code, Cowork, Chrome)	Must-have
LinkedIn Sales Navigator	~$100	Prospecting	Must-have
Apollo.io	Free tier or ~$50+	Lead enrichment	High
Typefully (or Buffer)	Free or ~$13	Content scheduling	High
Canva Pro	~$13	Design	Medium
Shield Analytics	~$8	LinkedIn analytics	Nice-to-have
Google Search Console	Free	SEO tracking	Must-have
Plausible/Umami analytics	~$9	Website analytics	Recommended
Substack	Free	Newsletter	Must-have
Brevo	Free tier	Email automation	Must-have (existing)
TOTAL (Essential)	~$300-400/mo
TOTAL (Full Stack)	~$400-500/mo

The math: A traditional agency would charge $5 - 15 k / m o n t h f orco n t e n t + p ros p ec t in g + co m p e t i t i v e in t e ll i g e n ce . Y o u^{'} re g e tt in g e q u i v a l e n t o u tp u t f or$ 400-500/month in tools + ~8-10 hours/month of your time. That’s a 10-30x cost advantage — the exact Smartcuts multiplier you’re looking for.

Quick Reference Card

When to use Claude Code:

Batch content generation, blog deployment, data processing, script automation, multi-file operations, API integrations, sub-agent orchestration

When to use Cowork:

Document creation (docx/pptx/xlsx/pdf), scheduled tasks, plugin workflows, non-technical file work, voice-enforced content review

When to use Chrome:

Live web research, LinkedIn browsing, prospect triage, competitor analysis, form filling, browser-to-doc handoff, content mining from feeds

The compound effect: Content → Outreach → Intelligence → Referrals → Delivery → Repeat Each layer feeds the next. Claude automates the labor. You provide the judgment and relationships.

This playbook is a living document. Update it as you test workflows, discover what converts, and as Claude’s capabilities expand.

Sources Referenced:

Solanasis Docs

Explorer

solanasis-claude-code-cowork-gtm-master-playbook