agentic-income
Skill2 starsChanged 5 months agoMITView on GitHub
The operating brain for building income systems with AI agents. Use when planning, building, or scaling an affiliate/content/product income network — deciding what to build next, where money actually comes from, how to make it compound, and how the system improves itself. Composes affiliate-audit. Trigger phrases: build an income system, make money with agents, scale my content network, what should I build next, passive income with AI, monetize this site, agentic income.
3 other projects use a version of this, 2 of them edited.
Adopted by
Most recently taken up by
- Starlight-Intelligence/starlight-agent-skillsedited, first seen Nov 8, 2026
- frankxai/agentic-income-skillstheir own version, first seen Oct 2, 2026
- frankxai/starlight-agent-skillsedited, first seen Jul 27, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 2
- frankxai/starlight-agent-skills★ 167% same
- Starlight-Intelligence/starlight-agent-skills★ 167% same
Wrote their own version · 1
- frankxai/agentic-income-skills★ 5same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository4 other files it gives its agents.
- AGENTS.mdAGENTS.md
- llms.txtllms.txt
- affiliate-auditSkill
- affiliate-opsSkill
The file
We split this skill into the rules it gives an agent
4 rules. 4 are common, 0 are distinctive, and 0 are very specific to this project.
4
Common
Most of the 3 edited versions we read keep these.
0
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Common: line 19
Own the audience: Every reader is income *or* a future relationship.
- Common: line 49
When asked "what should I build/do next," rank candidate actions by leverage:
- Common: line 53
Fork a spoke: only once the hub has ≥1 ranking post proving the shape works.
- Common: line 56
Never: chase a one-time bounty over a recurring one, link a tool you haven't used, or build site N+1 while site N has un-linked payer …
“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
agentic-income
The brain that turns "make money with AI agents" into a system that scales itself. One thesis, five principles, four loops. Everything else is execution.
The thesis
Honest AI-tool comparison content is the most scalable, lowest-cost income engine an AI architect can run. Frontier tools (ChatGPT, Claude, Midjourney, Veo) drive enormous search volume but pay nothing. So you rank for them, tell the genuine truth, and route to the adjacent tools that pay recurring — the ones you actually use. The reader gets the real answer; the system earns when they act on it. Trust is the asset; the link is downstream.
The five principles (non-negotiable)
- Honest pick always wins. Recommend what's genuinely best, then link the payer you'd actually buy. A link that overrides the truth burns the only asset that compounds: trust.
- Recurring > one-time. Passive income compounds on subscriptions, not flat bounties. Build around recurring-payers (Higgsfield, Systeme.io, CapCut, ElevenLabs, Copy.ai); treat one-time bounties as bonus.
- Own the audience. Every reader is income or a future relationship. One email capture per page turns rented traffic (SEO/social) into an owned list you control.
- Build the engine once, fork the sites. One shared catalog + one template; each new site is a
lib/site.tsswap. The second site is near-zero marginal cost. Effort goes into the substrate, not the Nth instance. - Compounding over spikes. Content that ranks for years + recurring commissions + a growing list = income that runs without you. Optimize for the asset that's still earning in 12 months, not the post that spikes today.
The architecture (hub-and-spoke)
affiliate-agent-skills ← the engine: catalog + audit + this brain (OSS — the lead-gen)
│
├─► agenticincome.ai ← HUB. The authority brand. Spokes link up to it.
├─► agenticpassiveincome.com ← spoke: the "set it and forget it" angle
└─► disruptivepassiveincome.com ← spoke: the "tools replacing job functions" angleOne engine, three brands, three search audiences, near-zero overlap. Cross-link spokes→hub for topical authority + referral flow. If one wins, double down; the losers cost almost nothing.
The four loops (this is what "agents that learn" means)
The system improves itself because each loop feeds the next. An agent runs these on a cadence — no human strategy meeting required.
1. Monetization loop (weekly): affiliate-audit joins the catalog × every site's content × traffic → ranks (a) which programs to join next, (b) which existing posts are mentioning a payer with no link. Join → set ourLink in data/programs.json → sync:catalog → links go live network-wide → re-audit, gap clears.
2. Content loop (weekly): pull each site's top-traffic + highest-intent queries → content-backlog.md ranks the next post by opportunity = authority × intent × monetization × low-competition. Build the top one in the citable shape. The winners tell you what to write next.
3. Authority loop (continuous): the OSS engine + public playbook earn GitHub stars and inbound links → that authority lifts the sites that consume the engine → more traffic → more audit signal. Giving the method away is the distribution.
4. Learning loop (monthly): every published post is a labeled example — query × shape × conversion. Feed outcomes back: which hooks/tables/answer-boxes converted, which programs actually paid. Bias the next batch toward what worked. The catalog's status + the backlog's ranking are the memory.
What to build next (the decision)
When asked "what should I build/do next," rank candidate actions by leverage:
- Set an
ourLinkfor a program already mentioned in a high-traffic post → instant revenue on existing traffic. (Runaffiliate-auditto find these.) Highest ROI, do first. - Write the top backlog post for the site with the most authority in that cluster.
- Fork a spoke only once the hub has ≥1 ranking post proving the shape works. Don't scale an unproven template.
- Join the next recurring-payer the audit flags as high-mention/no-link.
Never: chase a one-time bounty over a recurring one, link a tool you haven't used, or build site N+1 while site N has un-linked payer mentions.
The honest shape (every post)
Direct answer box up top (what AI search lifts) → sortable comparison table (the conversion surface; renders a CTA only when ourLink is set) → the genuine recommendation in prose → real FAQ with FAQPage JSON-LD → one affiliate disclosure. Recommend, don't sell. The shape ranks and converts.
Compose
affiliate-audit— the monetization-loop engine (catalog × content × traffic → gaps). This brain decides what to do; affiliate-audit finds where the money is.
Guardrails
- One disclosure per page with links (FTC + trust).
- Catalog
statusflags dead-ends (frontier LLMs) and closed programs — ignore them. - Re-verify affiliate terms before relying on them; they change often.
- Null
ourLink→ plain text, never a dead link.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

