agentleFS
erikunhaerikunha/vault-of-loci/.agents/skills/adaptive-scaffold/SKILL.md

adaptive-scaffold

Skill0 starsChanged 5 days agoMITView on GitHub

Read the record of an objective and teach it with a worked example and a similar problem when the learner is new to the topic or keeps failing on it, then drop the steps as the record rises; use it when the learner does not know a topic yet or keeps getting it wrong, and it writes a file, so run it only when asked.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the record of the objective.

    line 14
  2. Very specific:

    When you teach, follow the guess with a worked example from a source note, step by step.

    line 26
  3. Distinctive:

    When the learner gets it right, the steps come off: fewer steps, then no steps, then a new case.

    line 30
  4. 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 33
  5. Distinctive:

    Ask for confidence from 1 to 5 before the reveal.

    line 35

≈ “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 a guess comes before any help. 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.

The rule of the vault: teach a new learner, do not only test.

  1. Read the record of the objective. The report shows what was attempted and how it went, and the plan shows the item due for this objective.
python3 -I 90_System/Scripts/retention_report.py
python3 -I 90_System/Scripts/schedule_review.py --objective <id>
  1. Say in one line what the record says, so the learner can correct it. Then choose the level of support:
    • The record is empty, or the last attempt was wrong: teach.
    • The record shows correct attempts: ask straight away, with no worked example.
  2. When you teach, follow the guess with a worked example from a source note, step by step. Use the concept note in 20_Concepts and the sources it cites in 40_References. Then give a similar problem for the learner, and only then a second try. The learner works on the similar problem before the second try, not after.
  3. When the learner gets it right, the steps come off: fewer steps, then no steps, then a new case. What helps a beginner can hinder someone who already knows the topic. The scaffolding is dropped as the record rises and is never kept forever.
  4. 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.
  5. Ask for confidence from 1 to 5 before the reveal. Then reveal and grade the learner's words against the source.
  6. Record the attempt only after the learner confirms the result. Use the item id and the mode of the item. If the learner took a hint or a worked example, pass --assistance and --hints: an attempt with help stays out of the 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. Show the learner the line that was written. Do not edit the concept note and do not write the learner's study layer. 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.