agentleFS
Sign inStart free
multiples-hqmultiples-hq/business-state-machine-mcp/skills/call-to-action/SKILL.md

call-to-action

Skill2 starsChanged 47 days agoApache-2.0View on GitHub

Tell the person what the ledger says needs them - questions only they can answer, and follow-ups that are due - and offer to draft each follow-up. Use right after updating the state machine, or whenever the person asks what is waiting on them.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. charlie947/reel-editing-skills★ 9same 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.

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 7 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

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

    When the person asks: every chain. list_work lists all work newest first.

    line 12
  2. Very specific:

    If draft tools are on: open drafts from list_proposals.

    line 29
  3. Very specific:

    Read read_money_open once: for the whole business when the person asks, or with party_id for each customer or vendor the update touched.

    line 37
  4. Distinctive:

    Ask: chase now or wait?

    line 42
  5. Distinctive:

    If nothing needs them, say so in one line.

    line 66
  6. Distinctive:

    Draft a reply: write it from the ledger only: facts from the case brief and the evidence it cites, nothing the ledger does not show.

    line 72
  7. Distinctive:

    They handle it, or wait: leave the same memo, so the next run knows the follow-up is in hand and when to raise it again.

    line 78
  8. Distinctive:

    Never send anything. You draft.

    line 86
  9. Distinctive:

    Never put a fact in a draft that the ledger does not show, and never promise a date the evidence does not support.

    line 87
  10. Distinctive:

    Never ask the same question twice: check memos and drafts first.

    line 89

≈ “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.

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

Goal: the person knows, in a minute, what only they can unblock and which follow-ups are due, and each one has a next step they chose.

Which chains

  • After an update: the chains that update touched.
  • When the person asks: every chain. list_work lists all work newest first; read read_case_brief for each chain once (any ID in a chain gives the whole chain). Leave out a chain when every milestone of every work item is done and no latest memo has a waiting_for.

What to look for in each case brief

  • A judgment blocked whose explanation says the person must act, or that needs a fact only the person has.
  • A question the agent asked earlier that has no recorded answer: a waiting_for that starts with "Question for the person:". The brief shows each work item's latest memo and its waiting_for; read read_work_memos for a memo's details and for earlier memos.
  • A chain that follows no SOP (listed under not_recorded): the person must choose a version before anything is judged.
  • A latest memo whose next_review_at has passed, or a milestone blocked on someone outside the business: a follow-up is due.
  • If draft tools are on: open drafts from list_proposals.

A job moved to another SOP keeps the milestones its old procedure had under history_milestones, with their last status. They are history: leave them out.

Money

Read read_money_open once: for the whole business when the person asks, or with party_id for each customer or vendor the update touched. Take the amounts from it; never add them up yourself.

  • Owed to the business and late: who, how much, how many days past due. Ask: chase now or wait?
  • Owed by the business: who and by when.
  • A bill charge that is unassessed, or whose amount changed since it was judged: accept or dispute?
  • A payment with nothing allocated: what was it for?
  • Money paid before a bill, and a bill linked to a job with no lines yet: say so.

For one chain, read_work_money shows the agreed prices beside the bills: a vendor with an agreed price and no bill on the chain has not billed yet.

What to tell the person

Group what you found in two parts, most urgent first, one line of context each: the work's title, what is stuck, and the evidence behind it.

  1. Unblock me. Each question only the person can answer, asked exactly, with the choices when there are choices. One question per item.
  2. Follow-up needed. Who must act and on what. For each, ask how the person wants it handled:
    • "I draft a reply for you to send",
    • "you handle it yourself", or
    • "wait until a date you name".

If nothing needs them, say so in one line.

When the person answers

  • An answer to a question: hand it to update-the-state-machine, which records it as evidence and acts on it.
  • Draft a reply: write it from the ledger only: facts from the case brief and the evidence it cites, nothing the ledger does not show. Give it to the person, or to their mail tool as a draft if their harness has one. Then leave a memo on the work with leave_work_memo: follow-up drafted, waiting_for the reply, and next_review_at if they gave a date.
  • They handle it, or wait: leave the same memo, so the next run knows the follow-up is in hand and when to raise it again.

The reply, once someone sends it, comes back through the source as a new update. Nothing more to record until then.

Never

  • Never send anything. You draft; the person or their mail tool sends.
  • Never put a fact in a draft that the ledger does not show, and never promise a date the evidence does not support.
  • Never ask the same question twice: check memos and drafts first.

Done when: the person has seen every open item for the chains in scope, and each item they answered has its memo or its recorded answer.

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.

Check your setup, free