log-twitter-collab-maxxing
Skill63 starsChanged yesterdayMITView on GitHub
Collab maxxing: turn open conversations into collabs. A log play (docs/useCases/twitter/collabMaxxing.md): Network maxing gets people talking to the owner. Collab maxxing turns those conversations into something you can work on together. Everything the owner builds is open source and he earns nothing from it. So the collab is the deal: he builds their system with them and makes content about doing it, they get the outcome, and they back the work. Money is the last thing in the conversation, never the first. It comes as a small ask once they're keen, or a retainer the owner prices himself.
In this repository82 other files it gives its agents.
- AGENTS.mdAGENTS.md
- log-discord-channel-replySkill
- log-discord-forum-postsSkill
- log-discord-group-chat-readSkill
- log-discord-meme-replySkill
- log-discord-server-replySkill
- log-discord-server-searchSkill
- log-facebook-groupsSkill
- log-facebook-marketplace-listingsSkill
- log-inbox-boss-maxxingSkill
- log-inbox-governor-resetSkill
- log-inbox-reactionsSkill
- log-inbox-rea-looksmaxxingSkill
- log-inbox-scope-ajobSkill
- log-inbox-show-meSkill
- log-inbox-sign-inSkill
- log-inbox-triageSkill
- log-inbox-voice-noteSkill
- log-instagram-dm-replySkill
- log-instagram-new-platformSkill
- log-linkedin-networkSkill
- log-linkedin-postsSkill
- log-reddit-contributor-maxxingSkill
- log-slack-dm-replySkill
- log-twitter-affiliate-marketingSkill
- log-twitter-dm-inboxSkill
- log-twitter-engagement-repliesSkill
- log-twitter-engager-dmsSkill
- log-twitter-followersSkill
- log-twitter-follow-upsSkill
- log-twitter-get-the-actionSkill
- log-twitter-manager-maxxingSkill
- log-twitter-network-maxingSkill
- log-twitter-niche-map-regionSkill
- log-twitter-notificationsSkill
- log-twitter-recruitingSkill
- log-twitter-reply-guy-funnelSkill
- log-twitter-topic-repliesSkill
- log-twitter-who-needs-youSkill
- log-web-competitor-trafficSkill
- accountToDryRunSkill
- adsPowerSkill
- agentHooksSkill
- callTimingSkill
- captureRuleSkill
- commitLogSkill
- consoleSkill
- conversationsSkill
- devlogMaxxingSkill
- discordServersSkill
- facebookFlowsSkill
- facebookInspectSkill
- facebookWarmupSkill
- flowSkill
- governorSkill
- hostSkill
- humanMouseSkill
- inspectSkill
- instagramSkill
- knowledgeBaseSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
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
- Distinctive: line 47
Never ask the people the owner has ruled out for money (in the agent memory).
- Distinctive: line 48
Every send counts against the account's writes for the day, each line one write.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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.
Files
Collab maxxing: turn open conversations into collabs
This is the log play twitter/collabMaxxing. Before anything else, call the log MCP tool getGuide({ useCase: "twitter/collabMaxxing" }): 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).
Network maxing gets people talking to the owner. Collab maxxing turns those conversations into something you can work on together. Everything the owner builds is open source and he earns nothing from it. So the collab is the deal: he builds their system with them and makes content about doing it, they get the outcome, and they back the work. Money is the last thing in the conversation, never the first. It comes as a small ask once they're keen, or a retainer the owner prices himself.
The owner runs it across everyone already talking to him, the new people from networkMaxing.md, the friends with open threads and the missed openings nobody answered. On X it works the same in the old DM inbox and X Chat, and on Discord it works through discordRead and discordSendLive.
How the owner asks for it: "position a collab first then be like just a few quid so i can make a video about it", "start network maxxing and collab maxxing and trying to see what hes doing get him to reveal it all", "follow up on <someone> and the other people, small commitment from deals, not just new people but already open convos", "respond with the collabmaxxing", "selling the outcome", "let him name it", "check in with <someone> whats good there he is keen but isnt liquid", "just be nice to them there the poor people", "ask <someone> if i could become a contributor", "look on my twitter for missed opportunities".
Run it from a fresh context
Say: "go on docs/useCases/twitter/collabMaxxing.md", and say who it's for this round (everyone open, the missed openings, one person). The agent reads this page and README.md, loads getGuide({ topic: "xSending" }) (and getGuide({ topic: "discord" }) for Discord), and runs the steps.
The steps
| # | Step | Tool | Done when |
| 1 | Who: everyone waiting on the owner and everyone who opened his last message, plus the missed openings. The missed openings are threads in the local store where they wrote last, or said something with a deal in it (hiring, a retainer, a price, "lets do it", an intro), and nobody answered. Keep the real people. Drop spam, cold pitches at the owner, group banter and notification bots | twitterDms { account, waiting: true, limit: 200 }, search { query }, discordChats | a short list: who, what they said last, and what the opening is |
| 2 | Read each one's full context right now: the whole thread (old DM and X Chat merged) and what they did on the owner's posts. On Discord, read the chat | twitterDms { account, with: "<handle>" }, discordRead | what they do, what they asked, and where the talk stands |
| 3 | Sort each person: keen (said yes to working together), scoping (has a need, hasn't said what), skint (said they can't pay), friend (crew, not a lead), cold (never talked). Each gets a different next message | the agent | one bucket per person |
| 4 | Draft the one next move for their bucket, in the owner's words for that person. Scoping: get them talking about what they do, how they sell and what they spend, with one question and often none. Ask what it looks like in 6 months. Keen: position the collab: it's all open source and free, you get the outcome, I make content about us doing it, would be lit to properly collab. Let them lead on what's first. Skint: be nice, no pressure, ask nothing about money, and leave the door open. Friend: react to what they said and share something of the owner's, never a script. Cold: the opener only (networkMaxing.md step 10) | voice { to, with, context } | one draft per person |
| 5 | Gate each draft against the same person, and rewrite until it passes. A draft with a price, a retainer figure or a meeting waits for the owner: say what it would offer and let him give the number | voice { to, with, context, text } | the gate passes, or it's held for the owner with why |
| 6 | Send. One person: twitterDmLive. Two or more: the account's one queue, never a loop. On Discord: discordSendLive. The recipient is the numeric X user id. A handle is refused (userIdInvalid), so look it up first with treg's X profile read | twitterDmLive { userId, text, approvedBy, confirmLive: true }, or twitterQueueAdd and twitterQueueRunLive, or discordSendLive | the platform's answer is sent |
| 7 | When they answer, answer straight away if it fits the conversation. When they're keen, the ask comes next: a few quid so the owner can make a video about it, or the retainer he priced. Then tell the owner what they said and what went out | twitterDms { with }, voice, twitterDmLive | the owner knows where each collab stands |
What to know
- The pitch, in the owner's words: he builds open source and earns nothing from it, so collab so he can make a few quid, they get the outcome and he makes content about it. Frame it as joining the crew: his systems find, warm and book people, and someone still has to be on the other end of the call.
- Sell the outcome, not the tool. Ask what they'd want in 6 months and how it should feel. "Let him name it" means let them say what it's worth before the owner sets a number. A price or retainer is always the owner's call. Once, a contact's retainer was held at the owner's own figure after it came out what that contact charges their own clients.
- Learn the model from people who run it. A contact who runs an agency once explained theirs: a big retainer up front, then a smaller retainer plus rev share. Some clients pay a small retainer plus rev share from day one, because the agency owns most of their sales process. Answer people who explain things like that, because it's how the owner prices his own collabs.
- Never run the script on a friend. A friend once worked out the chat was automated. What gave it away: a discovery-question ladder in 15 minutes, money asked before they raised it, every reply 40 to 90 s after their message and ending on a question, and a queued line landing over their question. With friends: one question at most, vary the timing and length, never send while their newer message is unanswered, and hand money talk to the owner.
- Skint is not a no. When someone says they can't pay, don't push for a figure. One contact was "keen but isnt liquid". The check-in said no rush, when the money's there it's there, still keen to build something, and they came back keen and offered to make content on it for their own market. Another said they were in survive mode, and got a kind message and nothing asked.
- Collabs go both ways. Asking to contribute to someone else's open source counts too. The owner once asked another open source maintainer to be a contributor, opening on the maintainer's own reply to his post.
- Missed openings are the cheapest wins. Someone who explained their model, said they're hiring, or said "lets do it", and never got an answer. Find them in step 1 before reaching for new people.
- Never ask the people the owner has ruled out for money (in the agent memory). Never send the same message twice: check the thread before a resend. Memes go as media, never as a raw link.
- Every send counts against the account's writes for the day, each line one write. Space people out and never act in quick succession. Any refusal stops the run (rule 94).
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.

