fsbtactic-code/marketing-brain-skill / skills/marketing-brain
marketing-brain
FlaggedCreate, improve, research, or audit marketing strategy and execution using a bundled marketing library, verified live primary sources for volatile facts, and clearly labeled agent judgment. Use for full marketing systems, audits, landing pages and CRO, competitor intelligence, brand voice, audience and JTBD research, email sequences, SMM and social content systems, paid ads, headlines, copywriting, positioning, offers, launches, GTM, growth, funnels, retention, pricing, SEO, analytics, attribution, experiments, and consumer or buyer psychology. Trigger for English and Russian requests, including requests for ready-to-use assets.
references/workflow-social.md
View on GitHub# Social and SMM Workflow
Use this workflow to create an evidence-grounded social strategy, content system, calendar, or production-ready asset pack.
## Trigger conditions
- The user asks for SMM strategy, social positioning, channel selection, or a content operating system.
- The user asks for a social calendar, posts, carousels, short-video scripts, stories, threads, or community prompts.
- The user asks to audit current accounts or improve social performance.
- The user asks to repurpose an existing article, video, podcast, webinar, launch, or research asset.
Do not use this workflow for a mechanical rewrite of one supplied post unless the task introduces new claims or strategy.
## Primary output
Create `SOCIAL-CALENDAR.md` when file output is requested or artifact mode is active. Otherwise return the same contract in the response.
The first line after the title must be `Workflow status: PASS`, `Workflow status: NEEDS_DATA`, or `Workflow status: BLOCKED`.
## Input contract
Required:
- Brand, product, real offer, market, language, and business outcome.
- Priority audience, customer state, relevant job, and channel behavior evidence.
- Available accounts or candidate platforms and the role social should play.
- Brand voice, approved claims, CTA destinations, and content exclusions.
- Production capacity, approval owner, period, and requested asset depth.
Useful but optional:
- Account exports with post-level reach, watch, save, share, comment, click, and conversion data.
- Existing posts, creative assets, customer questions, interviews, sales calls, and support themes.
- Campaign calendar, launches, seasonal constraints, partnerships, and paid amplification plans.
- Rights-cleared visual assets, spokesperson availability, and editing capabilities.
- Current event taxonomy, link tracking, attribution limits, and reporting cadence.
Ask only for missing inputs that change channel choice, claims, production feasibility, rights, or measurement.
## Dependencies
- Use user account data before generic creator advice.
- Load `BRAND-VOICE.md`, `AUDIENCE-RESEARCH.md`, and relevant product artifacts when present.
- Run retrieval before choosing behavioral, persuasion, content, or community mechanics.
- Verify platform facts and policies live from official sources.
- Treat trends, creator examples, and legacy platform advice as discovery leads, not durable rules.
## Retrieval plan
Run two or three focused queries adapted to the audience and format:
```text
python scripts/retrieve_marketing_context.py --query "social content audience job sharing saving community trust" --domain organic-smm --limit 5 --json
python scripts/retrieve_marketing_context.py --query "editorial series hook narrative repurposing content quality" --domain content-editorial --limit 5 --json
python scripts/retrieve_marketing_context.py --query "social measurement watch time conversion experiment decision rule" --domain measurement-analytics --limit 5 --include-discovery --json
```
Use `creator-community` for community systems and `consumer-behavior` for a specific decision question. Open canonical discovery sources before relying on them.
## Live official checks
- Verify current format support, duration, aspect ratio, text fields, link behavior, collaboration features, and account requirements in official platform documentation.
- Verify current advertising, disclosure, branded content, contest, commerce, moderation, and regulated-content rules when relevant.
- Verify music, image, footage, testimonial, and user-generated content rights before recommending publication.
- Verify current product facts, prices, dates, availability, and destination links on canonical pages.
- Use account analytics or an explicit test for timing and cadence. Do not declare a universal best time or posting frequency.
- Record the observed date for every changing platform or trend claim.
Do not present an algorithm rumor, hashtag count, engagement trick, or current trend as a stable fact without official evidence.
## Execution phases
### Phase 1: Normalize the social brief
Define the business decision, audience, customer state, channel role, offer, voice, production capacity, period, and success event. Mark assumptions and unresolved constraints.
### Phase 2: Audit available evidence
Inspect current account data and content when available. Identify recurring topics, formats, audience questions, conversion paths, production costs, and evidence gaps. Do not infer causality from top posts alone.
### Phase 3: Select channel roles
Choose platforms only when audience behavior, product fit, capabilities, and production capacity support them. Give each selected platform one primary role and define why other platforms are deferred.
### Phase 4: Build the editorial spine
Define three to five content pillars, recurring series, audience questions, proof sources, narrative boundaries, and conversion role. Each pillar must connect to a customer job and business outcome.
### Phase 5: Design the format system
Map each idea to a native format using currently verified capabilities. Define hook, content promise, body or script, proof, visual structure, CTA, destination, accessibility, rights, and reuse path.
### Phase 6: Build the calendar
Sequence content around audience learning and business priorities, not arbitrary percentages. Use dates, windows, and time zones supported by account data or mark timing as a test.
### Phase 7: Produce complete assets
For every calendar item marked `ready`, write the full post, thread, carousel copy, or spoken script plus the visual or shot brief. If the user requests a complete calendar, every scheduled item must be `ready`.
### Phase 8: Design community operations
Define response ownership, escalation, moderation, saved replies, question capture, customer-care routing, disclosure, and response windows that the team can actually support. Do not prescribe engagement pods or coordinated inauthentic activity.
### Phase 9: Design repurposing
Create a source-to-asset map. Preserve the original claim, context, rights, and CTA while adapting structure to each platform. Avoid duplicating identical copy across channels.
### Phase 10: Define measurement and learning
Create a KPI tree from exposure to customer progress and business outcome. Define denominators, observation windows, tagging, qualitative signals, review cadence, and decisions to scale, revise, or stop.
For tests, define assignment unit, changed variable, primary metric, guardrails, and decision rule. Do not invent a universal benchmark, sample size, or expected lift.
## SOCIAL-CALENDAR.md contract
Use these headings in this order:
1. Executive social decision
2. Workflow status and blocking data
3. Business context, audience, and assumptions
4. Evidence from current accounts and research
5. Selected platforms and channel roles
6. Editorial spine and recurring series
7. Format and creative system
8. Calendar
9. Complete production assets
10. Community and moderation playbook
11. Repurposing map
12. Measurement and learning loop
13. Production workflow, owners, and approvals
14. Evidence notes and open questions
For every calendar item include `ID`, `Date or test window`, `Platform`, `Account`, `Audience state`, `Pillar`, `Goal`, `Format`, `Hook`, `Full copy or script`, `Visual or shot brief`, `Primary CTA`, `Destination`, `Proof source`, `Rights or disclosure`, `Owner`, `Metric`, and `Asset status`.
## Quality gate status
- `PASS`: the audience and channel roles are supported, changing platform facts are verified, every scheduled ready asset is complete, claims and rights are clear, production is feasible, and measurement can inform a decision.
- `NEEDS_DATA`: a safe partial strategy is possible, but account data, product proof, voice approval, asset rights, destination, production capacity, or measurement inputs are missing. List each missing item and affected asset.
- `BLOCKED`: publication depends on fabricated proof, deceptive engagement, impersonation, undisclosed sponsorship, unlicensed material, prohibited content, unsafe regulated claims, or another defect that cannot be repaired within the brief.
Assign exactly one status. A full-looking calendar is not `PASS` when assets, rights, owners, or tracking are unresolved.
## Stop and fallback rules
- Stop channel selection if the audience and business outcome are unknown.
- Stop production copy if product facts, offer, voice, or CTA destination are unknown.
- If account analytics are absent, label timing, cadence, and format mix as tests.
- If visual inspection is unavailable, avoid claims about composition or current account design.
- If trend verification is unavailable, use evergreen formats and leave trend slots unscheduled.
- If rights are unclear, provide an original production brief and exclude the disputed asset.
- If the requested volume exceeds production capacity, narrow the calendar and mark the gap `NEEDS_DATA` rather than lowering asset quality.
## Completion criteria
The workflow is complete only when every channel has a defensible role, every content pillar serves a customer job, every scheduled ready item is production-complete, claims and rights are controlled, community ownership is explicit, measurement drives a decision, the assigned status is justified, and no stale platform assumption is presented as current truth.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |