AgentStack
SKILL verified MIT Self-run

Create Marketing

skill-bnet47-codexicon-create-marketing · by bnet47

Create evidence-based positioning, copy, campaigns, and launch assets. Use for marketing strategy or execution, conversion, channels, or measurement; never invent proof.

No reviews yet
0 installs
0 views
view→install

Install

$ agentstack add skill-bnet47-codexicon-create-marketing

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README — it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-bnet47-codexicon-create-marketing)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
2d ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming — see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps — measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Create Marketing? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Create marketing

Produce marketing that is specific, credible, and useful to a real audience. Ground persuasion in product truth and customer evidence.

1. Load product truth

Read the approved charter, relevant briefs, real product behavior, existing brand guidance, and current customer-facing material. Read agent_docs/marketing.md when it exists; create or update it only when the request includes durable marketing context.

Resolve the minimum missing context:

  • audience, segment, buying or adoption situation, and awareness level;
  • problem, current alternative, trigger, and cost of inaction;
  • product, offer, differentiation, limits, and availability;
  • objective, journey stage, channel, format, and one primary action;
  • brand voice and words to use or avoid;
  • available proof, objections, legal constraints, and measurement.

Maintain an evidence ledger while working:

| Item | Status | |---|---| | Product behavior or claim | sourced, user-supplied, inferred, or unknown | | Customer language | named interview, review, support, sales, or unknown | | Market or competitor fact | current primary source, dated secondary source, or unknown | | Proof | approved, needs approval, or unavailable |

Never silently promote an inference into copy. Research current markets, competitors, channel rules, pricing, or benchmarks from primary sources when they materially affect the work.

2. Route only relevant detail

  • For pages, offers, copy, conversion, forms, onboarding, or pricing presentation, read [references/copy-and-conversion.md](references/copy-and-conversion.md).
  • For launches, email, social, paid media, partnerships, PR, or multi-channel campaigns, read [references/campaigns-and-channels.md](references/campaigns-and-channels.md).
  • For customer research, positioning, SEO discovery, analytics, experiments, or strategy, read [references/research-and-measurement.md](references/research-and-measurement.md).

Load only the references the task needs.

3. Set the argument before assets

Define:

  1. one audience and situation;
  2. one important problem or desire;
  3. one clear promise or value exchange;
  4. one credible reason to believe;
  5. one main objection to resolve;
  6. one next action;
  7. one measurement that would change a decision.

If audiences, awareness levels, or channels require different arguments, separate them. Do not flatten incompatible needs into generic copy.

4. Create channel-native work

  • Lead with the audience's situation, not company adjectives.
  • Translate features into concrete consequences without overstating outcomes.
  • Use real customer language only from a named source.
  • Match length, structure, pacing, format, and action to the channel.
  • Prefer one strong concept executed consistently over disconnected ideas.
  • Generate variants only for meaningfully different hypotheses; label each angle and what it tests.
  • Preserve a stable promise, proof, voice, and offer across channel adaptations.
  • Use specialist document, presentation, image, video, site, or spreadsheet workflows for the actual artifact and verify rendered output.

Reject vague transformation promises, buzzword stacks, interchangeable formulas, fake urgency, invented scarcity, unverified superlatives, implied guarantees, fabricated proof, confident unsourced market facts, and decorative content that does not advance understanding or action.

5. Protect truth, people, and permission

Preserve qualifications that are legally, scientifically, financially, or operationally material. Flag claims requiring owner, legal, compliance, privacy, accessibility, or subject-matter approval. Do not target or personalize using sensitive attributes, deceptive tactics, or data without a legitimate documented basis.

External publication, outreach, ad activation, account changes, spend, or messages require explicit authorization. Drafting does not authorize sending.

6. Review and hand off

Check clarity, relevance, specificity, differentiation, credibility, voice, channel fit, accessibility, and measurement. Confirm the work still makes sense when unsupported proof is removed.

Deliver:

  • the finished asset or strategy;
  • audience, argument, channel, and hypothesis;
  • sources and approved proof used;
  • assumptions, unknowns, and approvals still needed;
  • measurement event, metric, decision rule, and instrumentation gap;
  • what was rendered, tested, or not validated.

Use $review-creative for an independent pass on important customer-facing work.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

  • Author: bnet47
  • Source: bnet47/codexicon
  • License: MIT
  • Homepage: https://bnet47.github.io/codexicon/repo-template-playbook.html

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.