Mb Organic

Organic

Create organic content scripts in your voice — Reels, TikToks, carousels, static posts.

Shared source: The portable workflow contract lives in workflows/mb-organic/workflow.md. This Claude skill is the Claude Code shell over that source.

Shared contract markers: Keep these aligned with the shared source.

Required commands:

• mb status --json --peek
• mb start --json
• mb doctor repair --plan
• mb validate --cross-refs --json
• mb checkpoint --plan --json

Required fact paths:

• money_path
• money_path.objects.proof.quality
• money_path.objects.channel_strategy
• money_path.objects.active_push
• validation.file_contracts
• content_strategy
• content_strategy.overall_state
• content_strategy.simple_entry_point
• content_strategy.layers
• content_strategy.layers.distribution
• content_strategy.layers.channels
• content_strategy.layers.accounts
• content_strategy.layers.people
• content_strategy.findings
• ranked_actions
• update
• readiness
• drift.items
• relationship_health.gaps
• checkpoint.pending
• checkpoint.pending.blockers
• runtime.codex_cli
• runtime.claude_code

Approval gates: updates_repairs_migrations, file_writes, checkpoint, provider_mutation, publishing_or_spend, customer_contact, private_data, destructive_operations, structured_collection, and public_issue_or_proposal.

Public/private boundaries: no_secrets, no_raw_provider_exports, no_raw_transcripts, no_customer_member_data, no_private_runtime_settings, no_private_dms_or_gated_communities, and no_raw_finance_legal_records.

Core flow: support plan, video, carousel, static, sales-video-repurpose, or review. Marker phrase: plan, video, carousel, static, sales-video-repurpose, or review. If the operator wants mining or source collection, route to mb-think. Use content_strategy.overall_state and money_path.objects.proof.quality before drafting; keep source/privacy boundaries visible; route coordinated drafts to pushes/<YYYY-MM-DD-slug>/organic-batch-001.md; Do not publish, schedule, upload, mutate provider accounts, spend, auto-DM, auto-reply, or contact customers. Codex support stays read-only planning until runtime smoke proves organic drafting writes.

Need help? Type /mb-help + your question anytime. If conversation compacts, /mb-help reloads fresh context.

CLI facts first: Run mb status --json --peek from the business repo before triage, reference-health checks, or content planning. Use its readiness, drift, and ranked-action facts instead of hand-rolled repo-health probes.

Output destinations and operator vocabulary

When this skill produces a coordinated push (a content sequence with a goal, audience, and review date), write the wrapping record to pushes/<YYYY-MM-DD-slug>/push.md (type: push, linked_pushes for inbound links). One-off scripts and source captures route to documents/transcripts/ or documents/prototypes/ per the artifact-routing rules in the system architecture doc; they don't need a push wrapper.

If a content push uses a comment-keyword, DM-keyword, reply/link, resource delivery, or external provider setup plan, write the durable plan to pushes/<YYYY-MM-DD-slug>/playbooks/<playbook>.md with type: playbook. The playbook records the plan, approval state, safe provider state, validation evidence, and outcomes. It does not execute provider mutation.

If core/vocabulary.md defines display words (e.g. terms.push.singular: drop), speak the operator's word in conversation while still writing engine files. If the repo still has legacy campaigns/ records, recommend mb doctor and mb migrate campaigns --plan before creating new push work.

When creating push.md, include the validator-required frontmatter and fill the values from repo truth or operator answers:

---
type: push
slug: YYYY-MM-DD-slug
kind: nurture
status: planned
health: unknown
goal: { metric: "", target: "", by: YYYY-MM-DD }
owner: ""
audience: ""
offer: core/offers/<offer>/offer.md
promise: ""
---

If the push is tied to a bet, decision, research file, playbook, or outcome, add the appropriate typed frontmatter link (linked_bets, linked_decisions, linked_research, linked_playbooks, linked_outcomes). Mirror frontmatter links in ## Related links with Markdown relative links, or preview mb doctor repair --plan and ask before applying the repair. Use the connection decision matrix in docs/business-connections.md before adding typed links. Do not infer frontmatter links from body-only references.

Triage

Detect if user is in the right place:

If mining intent detected:

"Sounds like you want to research/mine competitor content. That's /mb-think territory — it saves to research/ and feeds your reference files. Should I switch you to /mb-think?"

If unclear:

"Are you trying to mine competitor content (research) or create new scripts (generate)?"

Pull Latest Updates

For the standard Main Branch update check and failure warning, see references/pull-engine-updates.md. Run it at the start of every invocation.

Then run mb status --json --peek from the business repo and use its readiness, drift.items, and ranked_actions facts before asking setup or reference-health questions. Direct file checks below are content-specific, not repo-health probes.

About Mining (Lives in /mb-think Now)

Mining competitor content is research work. It belongs in /mb-think because:

• Mining output goes to research/ folder
• Extracted insights feed into reference files
• It's the "Research" phase of the think cycle

If user wants to mine: Route them to /mb-think. Say:

"Mining is research work — /mb-think handles that and saves to your research/ folder. Should I switch you over?"

This skill assumes mining already happened. Users arrive here with:

• Concepts from research/*-competitor-mine.md
• A topic they want to turn into content
• Inspiration from their own research

If a selected research file has brief_format: grok-8, use its downstream handoff before generating: ICP and audience language for hooks, journey stage for format and CTA, competitive landscape for contrast, brand story for voice, content/assets for proof, and metrics/constraints for the success bar. If the brief names a resource-delivery mechanic, write or update a push playbook as a plan only; do not execute provider mutation.

When organic work is meant to sharpen an offer instead of only create content, load .claude/reference/conversion/offer-sharpening.md. Use the rubric to turn audience language, objections, proof angles, and soft CTAs into durable offer updates through /mb-think; do not treat comments or saves as buyer proof by themselves.

For the full mining methodology (Visual/Audible/Emotional framework, AI capabilities and limits, why mining flows into reference) see references/mining-methodology.md.

Offer Context Resolution

Before loading reference files, resolve active offer context with .claude/reference/business-primitives/offer-bet-push-proof.md:

1. If a future mb JSON field exposes active offer state, use it.
2. Do not treat .vip/local.yaml as the source of truth for active-offer state. If legacy state exists, confirm the offer with the user instead of silently routing.
3. If an offer is selected and core/offers/[offer]/offer.md exists, load it as the active offer.
4. If no offer is selected AND core/offers/ exists: ask which offer.
5. If no core/offers/ folder: use core/offer.md (single-offer mode)
6. If the repo does not have core/, do not infer current offer context from old paths. Run mb doctor repair --plan; use reference/* only as legacy migration input.

Always-core files: soul.md, voice.md, content-strategy.md Content strategy layers: read core/marketing/distribution-strategy.md, core/marketing/channels/<channel>.md, core/marketing/accounts/<platform>-<account>.md, and core/people/<person>.md when the run names a channel, account, founder, or weekly content plan that depends on them. Do not copy person voice into every account file; reference it. Offer-aware files: offer.md, audience.md Proof files: company-wide proof in core/proof/testimonials.md, core/proof/typicality.md, and core/proof/angles/; offer-specific proof in matching files under core/offers/[active]/proof/. Read older offer testimonial files as compatibility context only. Before turning proof into posts, videos, or outcome claims, read money_path.objects.proof.quality from mb status --json --peek. Treat generic testimonials, specific outcomes, offer linkage, typicality context, and outcome feedback as factual signals. Do not describe proof as good, high-converting, persuasive, or ready to win.

Offer argument: /mb-organic video [offer] "concept" — e.g., /mb-organic video community "morning routine" If offer specified, it selects the offer for this run only.

First-Time Setup

Requires core/voice.md (always core), plus resolved offer.md and audience.md (offer-aware — checks core/offers/[active]/ first, falls back to core/).

Optional but recommended: core/content-strategy.md — If present, /mb-organic reads the recognition target, pillars, asset jobs, non-publishing rules, and strategy links. If the repo has core/marketing/distribution-strategy.md, channel files, account files, or core/people/<person>.md, read the relevant ones before drafting. Works perfectly without them.

Congruence check: If core/operations/funnel/skool-surfaces.md exists, read it. Organic content should echo the same positioning and claims visible on the Skool about page and pricing cards. No contradictions between organic and the landing experience.

CWD-first: If current Main Branch markers exist in CWD, use it. If CWD looks like an old Main Branch repo, run mb status --json --peek and/or mb doctor repair --plan before saved config or discovery. Do not write to old repo structure. Otherwise, run mb status --json --peek and use its repo/readiness facts if available. If status cannot identify a repo, ask the user or run /mb-setup.

Missing files? See references/first-time-setup.md.

Presenting Options (Keep It Tight)

Don't list all modes in chunky blocks. Instead:

1. Check for existing concepts — Is there recent mining in research/?
2. Recommend ONE path based on their context
3. Mention alternatives briefly in one line

Example output (concepts exist):

Found recent mining (research/2026-01-20-competitor-mine.md) with 10 concepts.

Recommended: Pick a concept and generate a video script.

Other modes: `carousel "concept"`, `static "concept"`

Which concept interests you? Or provide your own topic.

Example output (no concepts):

No recent mining found. Two options:

1. Mine competitors first → `/mb-think` (saves to research/, come back here after)
2. Skip mining, give me a topic → I'll generate directly

Which works better for you?

Modes

/mb-organic (Default)

Check for existing mined concepts, pick one, generate scripts.

Check research/ -> Check content-strategy.md and relevant strategy layers (if present) -> Select concept -> Generate -> Output

If content-strategy.md exists and has pillars or asset jobs defined, suggest topics aligned to those pillars when the user has no specific concept in mind.

If no mining exists, prompt: "No mined concepts found. Want to mine competitors first? That's /mb-think — should I switch you over?"

/mb-organic mine (Routes to /mb-think)

If user types /mb-organic mine, redirect:

"Mining is research work now. Routing you to /mb-think for mining — it'll save to research/ and you can come back here to generate scripts from those concepts."

Then invoke /mb-think.

/mb-organic video "concept"

Generate Reels/TikTok script from a concept.

/mb-organic carousel "concept"

Generate multi-slide carousel copy from a concept.

/mb-organic static "concept"

Generate single-post caption from a concept.

Output path (all script modes): pushes/YYYY-MM-DD-organic-[offer]-{slug}/organic-batch-001.md (include offer slug in multi-offer mode; omit [offer]- in single-offer mode). On legacy repos that still have campaigns/, run mb migrate campaigns --plan first; do not write new content under campaigns/.

Output frontmatter:

---
type: output
format: video | carousel | static
date: YYYY-MM-DD
status: draft
platform: instagram | tiktok
---

Campaign name is REQUIRED. Ask user if not provided. Examples: january-hooks, transformation-series, pain-point-reels.

Context Awareness (Check Before Recommending)

At session start, scan what's been done:

1. Check research/*-competitor-mine.md — Who was mined? When?
2. Check pushes/*-organic-*/ — What scripts exist? Treat campaigns/ as a legacy fallback only.
3. Don't suggest re-mining same handles from today
4. Recommend generating from existing mining if concepts unused
5. Check core/content-strategy.md — What recognition target, pillars, asset jobs, and strategy links are defined?
6. If the channel/account/person is named, check the matching core/marketing/ and core/people/ files before drafting.

Example context-aware response:

Found today's mining (research/2026-01-20-competitor-mine.md):
- @cassie.schoonover, @likfoon already mined
- 10 concepts extracted, 2 scripts generated

Options:
1. Generate from remaining 8 concepts
2. Mine new competitors → `/mb-think`
3. Start fresh with your own topic

What should we call this batch? (e.g., "january-hooks", "transformation-reels")

Transparency

Before generating: show which reference files you're using. Before saving: show file paths.

Full Flow Walkthrough

1. Check for concepts — Look in research/*-competitor-mine.md or research/*-mining.md
2. If no concepts — Ask if they want to mine first (route to /mb-think) or provide a topic directly
3. Select concept — User picks from mined concepts or provides their own
4. Adapt to brand — Map concept to user's offer, audience, voice
5. Generate scripts — Use appropriate framework (video/carousel/static)
6. Save output — Scripts to pushes/YYYY-MM-DD-organic-{slug}/. Never campaigns/; that folder is legacy compatibility only.
7. Checkpoint prompt — run mb checkpoint --plan --json, show the proposed checkpoint and blockers, validate the chosen message with mb checkpoint --validate "..." --json, then after operator approval save with mb checkpoint --message "..." --yes.

Mining lives in /mb-think now. If user needs to mine competitors, route them there first.

Video Mode

Input: concept from mining, user topic, or research file.

Structure: Hook (0-3s) → Retain (3-45s) → Reward (final 5-15s)

See references/organic-frameworks.md and references/video-script-template.md.

Carousel Mode

7-10 slides: Hook → Value (one idea/slide) → Summary → CTA

See references/carousel-template.md.

Static Mode

Hook (first line) → Body → Soft CTA → Hashtags (optional)

See references/static-template.md.

Voice Adaptation

Read current voice context. If CWD looks like an old Main Branch repo, run mb status --json --peek and/or mb doctor repair --plan before relying on it. Match tone, use their vocabulary, avoid their "never say" list.

Authenticity: Sounds like creator (not copywriter). Uses contractions. Matches energy. No AI tells ("dive into", "unlock", "game-changer").

See references/organic-frameworks.md for soft CTA examples.

Integration with /mb-think

To save winning angles: route to /mb-think codify → core/proof/angles/.

Content Strategy Integration

If core/content-strategy.md exists, /mb-organic uses it to improve output:

• Pillar alignment: Suggest topics from defined pillars when user has no specific concept
• Asset job: Identify whether the piece should rank, teach, prove, compare, convert, announce, or start a conversation
• Platform format: Default to the format matching the target platform from channel or distribution strategy
• Account fit: Use account strategy for allowed topics, CTA path, cadence, and content mix when present
• Person voice: Use core/people/<person>.md for founder/person beliefs, stories, and fabrication boundaries when an account references it
• Content mix: Track which pillar types have been generated recently to maintain ratio balance
• Hooks library: Pull proven hooks from the hooks library section if populated
• Enemy-pillar mapping: Each pillar fights a named enemy. Content should position against the enemy. Check voice.md for declared enemies.
• Saves-first metrics: Weight save-ability highest when evaluating content concepts. Saveable = educational, actionable, reference-worthy.

If content strategy files do not exist, /mb-organic works exactly as before -- from mined concepts or user-provided topics. No warnings, no degradation.

Quality Checklist

Content: Hook stops scroll. One idea. Value before ask. Soft CTA.

Saves optimization: Is this content saveable? Educational, actionable, reference-worthy content drives saves. Saves are the #1 purchase intent signal — weight them above shares and likes.

Enemy clarity: Does this content fight a named enemy? Check voice.md for Named Enemies section. Enemy-driven content creates identity contrast.

Voice: Sounds like creator. Matches energy. Uses vocabulary. No AI tells.

Platform: Appropriate length. Correct structure. Optimized for retention/saves.

Skool congruence: If skool-surfaces.md exists, claims and positioning match live about page + pricing cards.

Examples & Compaction Recovery

For walkthrough examples (context-aware start, video-only, mining redirect) and the recovering-from-compaction protocol, see references/examples.md.

References

Setup: first-time-setup.md, minimal-voice-template.md, apify-setup.md

Frameworks: organic-frameworks.md

Sales video repurposing: sales-video-repurpose.md

Templates: mining-template.md, video-script-template.md, carousel-template.md, static-template.md

Methodology & Examples: mining-methodology.md, examples.md