agentleFS
matthewdonsemail-labmatthewdonsemail-lab/log/.claude/skills/log-twitter-who-needs-you/SKILL.md

log-twitter-who-needs-you

Skill63 starsChanged 2 days agoMITView on GitHub

Finding who needs something from you, or who matters. A log play (docs/useCases/twitter/whoNeedsYou.md): Go through the account's messages and find the people who clearly need something from the owner, or who matter: someone asking for work, a collaboration, a referral, a call; someone well placed (a fund, a big audience, a shared network). It reads what each conversation is about across its whole history, not just the last line, and says how close each one is to money. It sends nothing. What happens next with each person is the owner's call, and the reply itself is [followUps.md](followUps.md).

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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
  1. Distinctive:

    Read before judging: the last message alone is a label, not an answer.

    line 39

≈ “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

Finding who needs something from you, or who matters

This is the log play twitter/whoNeedsYou. Before anything else, call the log MCP tool getGuide({ useCase: "twitter/whoNeedsYou" }): 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).

Go through the account's messages and find the people who clearly need something from the owner, or who matter: someone asking for work, a collaboration, a referral, a call; someone well placed (a fund, a big audience, a shared network). It reads what each conversation is about across its whole history, not just the last line, and says how close each one is to money. It sends nothing. What happens next with each person is the owner's call, and the reply itself is followUps.md.

How the owner asks for it: "look through my messages and find someone who clearly needs something off me", "summarise everything on my twitter, how likely is it we make money", "who's worth following up".

Run it from a fresh context

Say: "go on docs/useCases/twitter/whoNeedsYou.md", and which account. The agent reads this page and README.md, loads getGuide({ topic: "xInbox" }), and runs the steps.

The steps

#StepToolDone when
1Who wrote last and is waiting: already in context at session start, read from the local inbox store, nothing fetchedthe SessionStart hookthe list is in context
2Bring the inbox up to date: the first read takes everything once, every read after only what is new (about 5 s)twitterDms { account, limit: 60 }every conversation, newest first, with who spoke last
3Shortlist: everyone waiting on the owner, plus anyone recent the owner wrote to who answeredtwitterDms { account, waiting: true, limit: 60 }a shortlist of handles
4Read each shortlisted person in full, old DM thread and X Chat merged, one person at a timetwitterDms { account, with: "<handle>" }their whole thread, oldest first
5What the conversation is about beyond the thread: earlier talk about the same thing anywheresearch { query, with: "<handle>" }the matches, with when
6Who they are: their bio, role and reach, and whether they're in Twentytreg's X profile read, Twentyone line per person
7Sort them, each with the line that shows it and its time: needs something from you (asked for work, a call, help, an answer), matters (role, reach, where they could take you), money (a concrete need the owner sells, a budget, a referral), warm (friendly, nothing asked yet), no action (pitches at the owner, spam)a table for the ownerevery person has a row and a suggested next move
8The owner says what to do with each one, in their own wordsthe ownerone instruction per person, then followUps.md

What to know

  • Read before judging: the last message alone is a label, not an answer. Quote the line that shows the need and its time ("asked for a call, 29 Oct 21:08").
  • Someone pitching the owner (selling ads, a service) is not a buyer. Say so.
  • "(could not decrypt)" means X Chat couldn't be opened for that message, not that the person said nothing. Never say "nobody replied" from it.
  • Reads go one person at a time. Never read the whole inbox in a loop of calls, and never in the same session as a bulk send (rule 94).
  • How likely money is gets said plainly, from what was said. Nothing is inferred that the person didn't say.

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.