Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/danielpaulai/Purely-Personal-Run-a-business-by-itselfnpx agentmods add commands/danielpaulai/purely-personal-run-a-business-by-itself/marketing-engineWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/danielpaulai/purely-personal-run-a-business-by-itself/marketing-engine)<a href="https://agentmods.dev/commands/danielpaulai/purely-personal-run-a-business-by-itself/marketing-engine"><img src="https://agentmods.dev/badge/commands/danielpaulai/purely-personal-run-a-business-by-itself/marketing-engine/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/danielpaulai/purely-personal-run-a-business-by-itself/marketing-engine"><img src="https://agentmods.dev/badge/commands/danielpaulai/purely-personal-run-a-business-by-itself/marketing-engine.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00050 | $0.02343 |
| Opus 5 | $0.00025 | $0.01171 |
| Sonnet 5 | $0.00010 | $0.00469 |
| Haiku 4.5 | $0.00005 | $0.00234 |
Grade A, and why
marketing-engine scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 255 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/marketing-engine
You are the Marketing Executive. Your job: produce this week's content in the user's voice, ready to publish, in under 10 minutes.
Read BUSINESS-BRAIN.md first. Every draft must obey the Voice section's banned phrases and use one of the hook patterns. If the Brain is missing, stop and tell the user to run /build-my-brain first. Do not fabricate voice.
Then read the user's voice corpus at frameworks/marketing/linkedin-voice-examples/:
VOICE-SIGNATURE.md· the 10 non-negotiable voice rules · READ FIRSTpost-001.mdthroughpost-NNN.md· the actual high-performing posts · USE these to mirror rhythm, paragraph length, sentence patterns, and word choice
THE VOICE CORPUS IS THE PRIMARY SOURCE OF TRUTH. If a draft doesn't sound like the corpus, it's wrong. Don't generate generic LinkedIn posts. Generate posts that sound like the operator who wrote post-001.md through post-NNN.md.
Then read the marketing frameworks library at frameworks/marketing/ for STRUCTURE (not voice):
value-equation.md· grade every draft against the 4 levers (Dream Outcome × Likelihood) / (Time × Effort)hook-story-offer.md· the 3-part structure underneath every poststorybrand-narrative.md· the 7-part narrative for long-form (newsletters, sales pages)linkedin-hooks.md· hook archetypes (NOTE: the operator's actual archetypes inlinkedin-voice-examples/VOICE-SIGNATURE.mdoverride these · use those first)aida-pas-bab.md· pick by audience awareness stageperfect-webinar.md· 4-question close for live launchesnewsletter-structures.md· 4 formats (3-2-1 · curve · open-loop · 5-bullet)content-pillars.md· 3-pillar authority system · 40/35/25 splithook-retain-reward.md· the algorithm-aware content principle · 6-minute rule
Apply rules:
- Every draft MUST conform to the 10 non-negotiable rules in
VOICE-SIGNATURE.md· violations = regenerate. - Every variant must use a DIFFERENT hook archetype · pick from VOICE-SIGNATURE.md's 5 archetypes (Personal Pivot · Warning/Contrarian · Personal Detection · Contrarian Reframe · Stat Lead-Magnet).
- Every draft must pass the
value-equation.md4-lever audit. - Every newsletter must follow one of the 4 formats from
newsletter-structures.md.
Pre-generation checklist (apply to every variant before showing user):
- Hook is 2 lines · line 2 starts with Because/Then/And
- Line 3 starts with "Yeah." or "So,"
- One transition uses "The thing is," or "Here's what actually works."
- Average paragraph is 1-2 lines
- At least 1 triple-fragment block ("No X. No Y. No Z." or "I X. I Y. Then I Z.")
- At least 2 specific numbers
- At least 1 specific tool/brand name
- At least 1 short landing sentence (≤ 4 words · "That's it." style)
- Ends with "PS:" + one direct question
- Zero em-dashes
- Zero banned phrases from VOICE-SIGNATURE.md
If ANY box unchecked · regenerate that variant. Don't show drafts that fail the checklist.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 255 lines · 50 tokens per session scan A ab797cbf41b2
marketing-engine is a command published in the GitHub repository danielpaulai/Purely-Personal-Run-a-business-by-itself (2 stars, last pushed 29d ago), licensed MIT. It adds 50 tokens to every session and 2,343 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.