/go-to-market

Go To Market — Research, Create, Distribute

One command to go from repo to live content across all channels. This command:

1. Researches your project and builds a unified brief (.social/project-brief.json)
2. Distributes content to Instagram, LinkedIn, and generates a pitch deck

The brief is the single source of truth — each channel reads from it instead of re-asking for context.

Set a runtime path helper once before running any script commands:

SOCIAL_ROOT="${CLAUDE_PLUGIN_ROOT:-social}"
if [[ ! -f "$SOCIAL_ROOT/scripts/generate_image.py" && -f "$SOCIAL_ROOT/social/scripts/generate_image.py" ]]; then
  SOCIAL_ROOT="$SOCIAL_ROOT/social"
fi

---

Phase 1: Research & Brief

Step 1: Locate the Project

• If the user provided an argument, check whether it's a local path or a GitHub URL.
  • If it's a GitHub URL, clone it to /tmp/research-repo and work from there.
  • If it's a local path, use that directory.
• If no argument was provided, use the current working directory.
• Verify the directory contains source code (look for README.md, package.json, pyproject.toml, go.mod, Cargo.toml, or similar).

Step 2: Extract Facts from the Repository

Read the following files (whichever exist) and extract project context:

• README.md (or README.rst, README.txt) — project name, description, features, usage
• package.json / pyproject.toml / go.mod / Cargo.toml / build.gradle — project name, dependencies, tech stack
• docs/ directory — any additional documentation
• Recent git log (git log --oneline -20) — team members, build velocity, recent activity
• Git contributors (git shortlog -sn --no-merges) — team composition

For each piece of information extracted, mark it as evidence (directly from repo artifacts).

Fill in these brief sections from repo data:

• meta.repo_url — from git remote
• project.name — from manifest or README
• project.description — from README first paragraph
• project.tech_stack — from dependencies and file types
• project.key_features — from README feature lists
• team.members — from git contributors
• narrative.traction.bullets — from commit count, recency, contributor count

Step 3: Check for Existing Brief

• Look for .social/project-brief.json in the project directory.
• If found, load it as the base. Do not overwrite fields that already have user-provided values — only fill in empty fields or add new evidence.
• If not found, start from the template at references/project-brief-template.json.

Step 4: Fill Gaps with User Input

Present what was auto-extracted in a summary, then use AskUserQuestion to fill in critical fields that couldn't be inferred:

Always ask (if not already filled):

1. "What problem does this project solve and for whom?" — fills narrative.problem, audience.target_users, audience.pain_points
2. "What's the hackathon context?" (name, theme, prize category, build time, team members) — fills hackathon.*
3. "What's the main angle or story you want to highlight in content?" — fills content_angles

Ask only if relevant and empty: 4. "What's your one-liner pitch?" — fills project.one_liner 5. "What stage is the project?" (hackathon / pre-seed / seed / growth) — fills project.stage 6. "Are you fundraising? If so, how much and what milestones?" — fills fundraising.*

Mark user-provided answers as evidence when they state facts, or assumptions when they express goals or projections.

Step 5: Optional Web Research

Ask the user: "Want me to research competitors, market size, and trends to strengthen your narrative?"

If yes:

• Use WebSearch to find competitors, market size estimates, and recent trends related to the project's domain.
• Add findings to narrative.market (market size, competitors) and narrative.why_now (trends, timing).
• Tag all web-sourced data as evidence with source URLs.

If no, skip this step.

Step 6: Save the Brief

1. Set meta.created_at to the current ISO timestamp (or keep existing if updating).
2. Set meta.updated_at to the current ISO timestamp.
3. Write the completed brief to .social/project-brief.json in the project directory.
4. Print a summary to the user:
   • Filled sections — list sections that have content
   • Gaps remaining — list sections still empty or marked as assumptions
   • Evidence vs. assumptions — count of each

---

Phase 2: Content Distribution

After the brief is saved, present the user with a distribution menu. They can pick which channels to activate, or run them all.

Step 7: Choose Distribution Channels

Use AskUserQuestion with multi-select:

"Which channels do you want to publish to?"

• Instagram — Generate media (AI image or video) + caption, post to Instagram
• LinkedIn — Write a professional post, publish to LinkedIn
• Pitch Deck — Generate an 8-slide investor pitch deck (Slidev → PDF)

The user can select one, two, or all three. Execute the selected channels in the order below. Between each channel, confirm with the user before proceeding to the next.

Step 8: Instagram (if selected)

Read .social/project-brief.json and use the instagram-content-strategy skill. Follow these steps:

1. Pick a content angle — Use content_angles from the brief. Ask which angle to use if multiple exist.

2. Choose media type — Ask the user:

   • Generate with AI → then ask Image (1:1, Imagen 4) or Video (9:16, Veo 3)
   • Provide my own → user supplies a file path

3. Generate media (if AI-generated):

   • Craft a generation prompt from project, narrative.problem, narrative.solution
   • For images:

     GEMINI_API_KEY="$GEMINI_API_KEY" python3 $SOCIAL_ROOT/scripts/generate_image.py \
       --prompt "<crafted prompt>" \
       --output /tmp/instagram_post_image.png \
       --model imagen \
       --aspect-ratio 1:1
     

   • For videos:

     GEMINI_API_KEY="$GEMINI_API_KEY" python3 $SOCIAL_ROOT/scripts/generate_video.py \
       --prompt "<crafted prompt>" \
       --output /tmp/instagram_post_video.mp4 \
       --model veo3 \
       --aspect-ratio 9:16 \
       --duration 8
     

4. Craft caption — Use brief context (project.one_liner, hackathon, narrative.problem.headline, meta.repo_url). Include hook, story, CTA, 15-20 hashtags.

5. Present draft to user for approval or edits.

6. Post — Use post_photo or post_reel MCP tools. Share result with Media ID and link.

Step 9: LinkedIn (if selected)

Read .social/project-brief.json and use the linkedin-cli skill. Follow these steps:

1. Pick a content angle — Use content_angles from the brief. Can use a different angle than Instagram.

2. Choose post format — Ask:

   • Text post (highest organic reach)
   • Text + Image (AI-generated or user-provided, 4:3 aspect ratio)

3. Generate image (if requested):

   GEMINI_API_KEY="$GEMINI_API_KEY" python3 $SOCIAL_ROOT/scripts/generate_image.py \
     --prompt "<crafted prompt>" \
     --output /tmp/linkedin_post_image.png \
     --model nanobanana-pro \
     --aspect-ratio 4:3
   

4. Craft LinkedIn post — Use brief context (project, audience, narrative.*, hackathon, meta.repo_url, meta.website). Follow LinkedIn best practices:

   • Hook in first line (shows before "...see more")
   • Short paragraphs, line breaks between ideas
   • Professional but human tone, first person
   • Under 1300 characters for best engagement
   • 3-5 hashtags only

5. Present draft to user for approval or edits.

6. Post — Use LinkedIn CLI:

   python3 $SOCIAL_ROOT/skills/linkedin-cli/scripts/linkedin_cli.py post create \
     --text "<the post text>" \
     --visibility public
   

Step 10: Pitch Deck (if selected)

Read .social/project-brief.json and use the repo-to-fundraising-pitchdeck skill. Follow these steps:

1. Read required references:

   • references/pitch-framework-borys-musielak.md
   • references/slidev-playbook.md

2. Map brief to 8-slide framework:

   • brief.project → Slide 1 (Company + one-line pitch)
   • brief.team → Slide 2 (Team)
   • brief.narrative.problem → Slide 3 (Problem)
   • brief.narrative.solution → Slide 4 (Solution)
   • brief.narrative.why_now → Slide 5 (Why now)
   • brief.narrative.traction → Slide 6 (Traction)
   • brief.narrative.market → Slide 7 (Market)
   • brief.fundraising → Slide 8 (Ask)

3. Generate Slidev deck:

   python3 $SOCIAL_ROOT/skills/repo-to-fundraising-pitchdeck/scripts/generate_slidev_pitchdeck.py \
     --brief .social/project-brief.json \
     --out slides.md \
     --repo-path .
   

4. Polish — Review the generated slides, ensure one idea per slide, add presenter notes with evidence/assumptions.

5. Export to PDF:

   slidev export slides.md --format pdf
   

6. Present the deck summary to the user. Offer to open for review or export to PPTX if needed.

---

Step 11: Wrap Up

After all selected channels are complete, print a summary:

• Brief: .social/project-brief.json (saved / updated)
• Instagram: posted (Media ID + link) or skipped
• LinkedIn: posted (post URL) or skipped
• Pitch Deck: generated (slides.md + PDF) or skipped

Remind the user they can re-run /go-to-market to update the brief and publish to additional channels.