agentleFS
agiea-brainagiea-brain/agency-skills/skills/paul-munro/pre-session-flash-card/SKILL.md

pre-session-flash-card

Skill1 starsChanged 16 days agoMITView on GitHub

Use this skill before every session with a recurring coaching or advisory client, when you need one page to run the hour from. Produces the flash card: why this hour matters, the calls you make before the session, the opening move with the words to say, what the client decides today, the development block they asked for, dates and owners, a run sheet, what is already settled, and one closing line on how to be in the room. Built from the transcript of the last session and the recap you sent. Optionally produces a pre-read beside it. Also use for "flash card", "session card", "run sheet", "prep card", "pre-read", "prep me for my next session", "coaching session prep".

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 12 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈12

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the transcript of the last session, the recap thread you sent, and the last debrief's notes on how the previous card held up.

    line 24
  2. Very specific:

    Write the sections in the fixed order in references/card-sections.md.

    line 26
  3. Very specific:

    Check every line against references/plain-language.md.

    line 33
  4. Distinctive:

    If a pre-read would help, build it after the card.

    line 34
  5. Distinctive:

    After the session, record in your debrief how the card held up: held, changed in the room, missed, overtaken, or not on the card.

    line 35
  6. Distinctive:

    MUST build the card from the transcript of the last session and the sent recap thread.

    line 47
  7. Distinctive:

    MUST give each decision that has to leave the session its own numbered line.

    line 48
  8. Distinctive:

    MUST tag every invented item PROPOSED until the coach rules.

    line 49
  9. Distinctive:

    MUST keep the numbers. Plain language simplifies the words around a number.

    line 50
  10. Distinctive:

    NEVER build the card from a debrief alone.

    line 51
  11. Distinctive:

    NEVER put outstanding admin on the card.

    line 52
  12. Distinctive:

    NEVER place the development block last in the run sheet.

    line 53
  13. Distinctive:

    NEVER add sections, including a "collect before you hang up" list.

    line 54
  14. Distinctive:

    NEVER show the card to the client.

    line 55

≈ “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

Applies before every session with a client you coach or advise on a recurring basis. Produces one page, the flash card, that you read while you talk and run the hour from.

The play

  1. Fix the session date and time, and name the source on the card: the calendar invite, your own words with their date, or inferred from the pattern. An inferred date puts the word "preliminary" in the header.
  2. Read the transcript of the last session, the recap thread you sent, and the last debrief's notes on how the previous card held up.
  3. Sort every candidate line with one test: does it need the client's judgment in real time, or only their action? Judgment goes in the room. Action goes in the follow-up email. Confirm any owed item is still owed before it reaches the card. Full rules: references/line-rules.md.
  4. Write the sections in the fixed order in references/card-sections.md. Anything you invented goes in "Decide these before you dial", tagged PROPOSED, in plain words: what it is, your recommendation, adopt or kill. The coach rules before the session.
  5. Build the run sheet from the protected block outward. Place the development block first or second, then fit the deadline items around it. The clock times cap your own items. Give the client one labelled slot of at least 20 minutes, placed where the transcripts show their material tends to arrive.
  6. Branch on what you know going in:
    • A line might resolve before the session: write its skip condition in the same line. "If it is in the folder by Thursday, skip."
    • A block rests on an event that may not have happened: add one line for that case. "If it did not happen: date it, move on."
    • An item has sat on two consecutive cards without being raised: move it to the recap email. It returns only if the client raises it.
    • The client sent their own agenda note, and you have agreed the hour is theirs to set: use the client-led order in references/card-sections.md.
  7. Check every line against references/plain-language.md. Then check length: one page preferred, two maximum, about 1,100 words as the ceiling. Cut whole items. Keep the sentence that says what each remaining thing is.
  8. If a pre-read would help, build it after the card. The card still runs the hour alone.
  9. After the session, record in your debrief how the card held up: held, changed in the room, missed, overtaken, or not on the card. The next card's cuts come from there.

What good looks like

  • First thing I check that most people skip: whether each owed item is still owed. Email is an incomplete record of what a client has received, so I check delivery, and ask where that channel is silent. A ledger that logs commitments without logging resolution manufactures fake debt.
  • The mediocre version is long and protects its best block with a note. One card ran about 1,900 words and its run sheet did not survive the first ten minutes. On three cards in a row the development block sat after the deadline items with a note to protect it, and ran last and short all three times. The slot is what protects it.
  • I know the card is right when I can run the hour from it with nothing else open, and every line needs the client's judgment in the room.
  • I know the session went well when the client's material fills the open slot. One ran about two thirds on what the client brought. That is the good outcome, and the card should be shaped for it.
  • Red flag: an item riding from card to card. One sat on two consecutive cards without being raised, and reached the client in three sentences of email instead. I was carrying it as a wish.

Rules

  • MUST build the card from the transcript of the last session and the sent recap thread.
  • MUST give each decision that has to leave the session its own numbered line. Nothing mandatory sits below the decisions block.
  • MUST tag every invented item PROPOSED until the coach rules. Then the line states the ruling and its date.
  • MUST keep the numbers. Plain language simplifies the words around a number.
  • NEVER build the card from a debrief alone.
  • NEVER put outstanding admin on the card.
  • NEVER place the development block last in the run sheet.
  • NEVER add sections, including a "collect before you hang up" list. That list was retired because it repeated the card's decisions as fragments with no context and slipped follow-up requests into the room.
  • NEVER show the card to the client.

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.