agentleFS
Sign inStart free
kleyver14kleyver14/study-skills/skills/study-close/SKILL.md

study-close

Skill0 starsChanged 43 days agoMITView on GitHub

Close the current session of the user's active study plan after its evaluation — asks the user to explain the core ideas in their own words and writes those as the session's notes, rewrites what is still weak, marks the session closed, and when the session ends a block writes the block closure summary (coverage, results, concepts, drift) ready to copy. Use whenever the user says "cerremos la sesión", "listo por hoy", "terminé", "guarda los apuntes", "close this session", or wants to wrap up a study session or a study week/block.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask: *"In your own words, what are the 2-3 ideas from this session you would explain to a colleague?"* One message, then wait.

    line 25
  2. Distinctive:

    Do not lecture: the point is their wording, corrected as little as possible.

    line 27
  3. Very specific:

    If external_followup: true in PLAN.md, also print that block as plain text and say it is ready to paste wherever they report.

    line 57
  4. Very specific:

    Run python3 $SS status "$PLAN" and say in one line where they stand and what /study-next will open.

    line 62

≈ “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

/study-close — close the session with the user's own words

Notes the model writes get read once. Notes in the learner's own words are the learning. This command makes the user say the ideas back (the practice that works even with no environment to touch) and records it where the next session, and the next Claude, will find it.

Step 0

SHARED="$HOME/.claude/skills/study-shared"; SS="$SHARED/scripts/study_state.py"
PLAN=$(python3 $SS registry resolve) || exit
python3 $SS next "$PLAN"

The current session must be evaluated, or studied with eval_type: none. Otherwise point to /study-eval (or /study-next if it is still pending).

Step 1 — explain it back

Ask: "In your own words, what are the 2-3 ideas from this session you would explain to a colleague?" One message, then wait. If an idea comes back wrong or vague, ask one follow-up on it. Do not lecture: the point is their wording, corrected as little as possible.

Step 2 — write

With the Edit tool, in the session file:

  • notes region: the user's ideas, lightly edited for clarity, kept in their voice. Add at most one line per idea if something essential was missing, marked as yours.
  • weak region: rewrite it (study-eval may have filled it after a failed attempt) with what is still not consolidated: failed or guessed concepts from this session's evaluations that are not consolidated in PLAN.md, plus anything the user says is still fuzzy. Each line: the concept, the correct idea in one sentence, the source link. These feed the next evaluations' review quota.
python3 $SS set "<session-file>" status=closed closed_date=$(python3 $SS today)
python3 $SS refresh "$PLAN"

Step 3 — if is_checkpoint: true, close the block

Append under the Block closures heading of PLAN.md:

### Block N — closed <date from `python3 $SS today`>
- Covered: sessions a–b (<areas>)
- Checkpoint result: <score> (<type>)
- Concepts consolidated: … · still pending: …
- Drift: <on track | N sessions behind | ahead>; buffer left <x/y>

If external_followup: true in PLAN.md, also print that block as plain text and say it is ready to paste wherever they report. Do not publish it anywhere yourself.

Step 4 — hand off

Run python3 $SS status "$PLAN" and say in one line where they stand and what /study-next will open.

Language: files in the plan's language, conversation in the user's (language.md). The notes keep the user's own words, untranslated.

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