agentleFS
matthewdonsemail-labmatthewdonsemail-lab/log/.claude/skills/log-twitter-affiliate-marketing/SKILL.md

log-twitter-affiliate-marketing

Skill63 starsChanged 2 days agoMITView on GitHub

Affiliate marketing through your DMs. A log play (docs/useCases/twitter/affiliateMarketing.md): You use a tool, you make content about it, and people message you about it. This page turns those conversations into affiliate income without sounding like a pitch: each person gets a reply about their own thread, an update on what you've built, and the tool with your affiliate credit only where it actually helps them. The other side of the page is the tool companies themselves: getting onto their partner, referral or creator program and booking the call.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Every DM counts as a write against the account's 240 a day.

    line 57

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Affiliate marketing through your DMs

This is the log play twitter/affiliateMarketing. Before anything else, call the log MCP tool getGuide({ useCase: "twitter/affiliateMarketing" }): it returns this page and makes it the session's play, so every log tool's answer says which step you are on and a send with no play loaded is refused (rules 157, 158). Then follow the steps in order.

Draft every message from that person's own thread and matt's real replies (voice with no text, search), never from memory, and never send two that read alike (sameAsSent).

You use a tool, you make content about it, and people message you about it. This page turns those conversations into affiliate income without sounding like a pitch: each person gets a reply about their own thread, an update on what you've built, and the tool with your affiliate credit only where it actually helps them. The other side of the page is the tool companies themselves: getting onto their partner, referral or creator program and booking the call.

Inbound and outbound are on other pages: dmInbox.md reads who is waiting, engagerDms.md opens new conversations.

How the owner asks for it: "answer the people asking about <tool> in my dms with my affiliate link", "turn the dms about <tool> into affiliate money", "follow up everyone who asked about the tool", "get me onto <company>'s affiliate program and book the call".

Run it from a fresh context

Say: "go on docs/useCases/twitter/affiliateMarketing.md for <tool>, link <affiliate link>". The agent then reads this page and README.md and runs the steps.

The steps

#StepToolDone when
0Session checktwitterSessionCheckloggedIn
1Read every conversation, X Chat included, and index itsearchIndex { account } (log search index)"N conversations, M messages" and the chunk count
2Find the conversations worth a follow-up: anyone who asked about the tool, offered work, a call or a collaboration, or went quiet on an open threadsearch { query: "<tool> enrichment", with, after } (several queries: the tool, "call", "consulting", "affiliate", "role")a short list of people and conversation ids
3Read each person's whole history before writing a word: the old DM thread and the X Chat thread merged into one timelinetwitterDms { account, with: "<handle>" }you know the last thing each side said, and what was promised
4Sort them: owes you an answer, you owe them something (a doc, a look, a reply), left hanging, competitor (building a rival tool), tool company (their partner program)a table for the ownerthe owner agrees the list
5Write one message per person: pick up exactly where the thread left off, one line on what you've built since, then the offer framed as something for them. The link goes only to people who don't use the tool yet, and never to a rival's founderhouse stylethe owner approves each line
6Tool companies: ask for the partner, referral or creator program, show the reach (views, who follows you), and offer two concrete call times in their mornings or a calendar linktwitterDm / twitterDmLivea time agreed
7Send, dry run first. Two or more people go through the queue as dmstwitterDm then twitterDmLive { userId, text, approvedBy, confirmLive }, or the README queue with action: "dm"X's answer is sent
8Watch for replies. The inbox hook flags a new message with its text, X Chat includedthe inbox watcher (agent hooks)each reply read with twitterDms { with } before answering

What a message looks like

hey @someone sorry this slipped past me. since we spoke i've been building <what you built> all in public <your site>. you were asking about <their question> so this might help, <tool> gives you <credit> free here <affiliate link>

hey <partner contact> the referral route works for me and i'm keen to hop on a call and plan it out. would <day> or <day> at <time> your time work, or send a calendar link and i'll grab a slot

The lessons that shaped it

  • Read the X Chat side first. A person's newer messages often live in X's encrypted Chat, not the old DM thread. Replying from the old thread made the account apologise for "sorry this got buried" to people it had already had calls with. twitterDms with with merges both; never answer without it.
  • If a reply missed the context, own it at once: one line of apology, the real context in your own words, then the actual follow-up (the doc you owe, the video they haven't watched, the role they already turned down).
  • Collaboration, not a pitch. The link is the last line, for people who'd use it. A rival's founder gets an honest "i'll try yours", never your link.
  • Partner programs move on proof. Views, signups you drove and who follows you (decision makers) beat follower counts. When the affiliate program is paused, ask for the referral fee and a call.
  • Keep the promises visible. Anything you said you'd send goes into the owner's list so it isn't lost the next time.

What to know

  • X Chat opens with the account's X Chat PIN, saved once in a masked terminal prompt (log twitter chat pin --account <account>). Without it, X Chat messages show who and when, not what.
  • Every DM counts as a write against the account's 240 a day. --force goes past it only when the owner says so.
  • A DM only lands if the person accepts DMs from accounts they don't follow.

Next

  • #35 tracks which affiliate link went to whom and which turned into a signup, per person, in Twenty.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.