agentleFS
erikunhaerikunha/vault-of-loci/.agents/skills/transfer-lab/SKILL.md

transfer-lab

Skill0 starsChanged 5 days agoMITView on GitHub

Write a transfer task for an objective that was studied but never applied, put it in the track's practice note with the item id and the rubric from the lab, and record the graded attempt in the learning ledger; use it when the learner wants to practise applying an idea instead of repeating it, and it writes a file, so run it only when asked.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    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:

    Run the lab and read the first objective that owes a task.

    line 12
  2. Distinctive:

    Read the objective line under "Tasks to write".

    line 18
  3. Very specific:

    Read the concept note and the sources it cites.

    line 21
  4. Distinctive:

    Write the task. It is a case the learner has not seen, built from the criterion and from the concept note and its sources.

    line 23
  5. Very specific:

    Put the task in the practice note of the track the concept belongs to: 10_Tracks/<NN>_<Track>/practice/Transfer_Lab.md.

    line 26
  6. Distinctive:

    If the answer is wrong, or the learner asks to try again, use the hint ladder one step at a time: hint, strong hint, partial example, …

    line 34
  7. Distinctive:

    Ask for confidence from 1 to 5 before the reveal.

    line 36

≈ “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

Tutor mode: the learner attempts first and without notes. The agent never writes the learner's notes or answers. Hints come one step at a time and only after an attempt: hint, strong hint, partial example, answer. Confidence from 1 to 5 is asked before the reveal. The agent never grades an attempt the learner did not confirm.

  1. Run the lab and read the first objective that owes a task.
python3 -I 90_System/Scripts/transfer_lab.py
  1. Read the objective line under "Tasks to write". It gives the concept note the objective belongs to, the criterion that decides what counts, the item id the new task must carry, the mode and the rubric with its three levels.
  2. Read the concept note and the sources it cites. The concept note is in 20_Concepts and the sources are in 40_References.
  3. Write the task. It is a case the learner has not seen, built from the criterion and from the concept note and its sources. Never build it from the example that was studied. The task asks the learner to apply the idea to the new case and to say why it applies.
  4. Put the task in the practice note of the track the concept belongs to: 10_Tracks/<NN>_<Track>/practice/Transfer_Lab.md. The track is the one the concept note lists in its tracks property. If the file does not exist yet, copy 90_System/Templates/Practice/Transfer_Lab.md into the practice folder of that track. Fill the Date, Objective and Task columns. Leave the Notes column empty for the learner. After the learner confirms the result, fill the Criterion met column with the level of the rubric.
  5. Give the task and wait. Say nothing about the content before the attempt: no excerpt, no summary, no hint.
  6. If the answer is wrong, or the learner asks to try again, use the hint ladder one step at a time: hint, strong hint, partial example, answer. Never skip a step.
  7. Ask for confidence from 1 to 5 before the reveal. Then reveal and grade the learner's words with the rubric the lab printed.
  8. Record the attempt only after the learner confirms the result. Use the item id and the mode the lab printed. If the learner took a hint, pass --assistance and --hints: a task with a hint stays out of the transfer rate.
python3 -I 90_System/Scripts/record_attempt.py --objective <id> --item <id> --mode <mode> \
  [--correct|--incorrect] [--at <iso>] --confidence <1-5> [--probability <0-1>] [--hints <n>] \
  [--assistance <level>] [--grading <method>] [--rubric <ver>] [--revision <rev>] [--minutes <n>]
  1. An attempt that repeats the definition instead of applying it does not count as a transfer. The rubric level is 0 and the objective still owes a task. Say that in one line.
  2. Show the learner the line that was written. Never edit or delete a line of the ledger.

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.