Skip to main content

your-skill-name

Use when <clear trigger condition for this skill>.

Zur Installation springen

Quellinformationen

Repository
yugash007/edu-agent-skills
Letzte Quellaktivität
18. Mai 2026 um 16:48
Erkannte Sprache von SKILL.md
Englisch
Sterne
7
Forks
2

Installationsoptionen

Standardmäßig ist der Prompt ausgewählt, der zuerst die Quelle prüft. Sie können zu einem direkten Befehl wechseln oder eine lokale Kopie herunterladen.

Quelldateien prüfen

Lesen Sie SKILL.md und alle von SkillsMP angezeigten Begleitdateien, bevor Sie sich für eine Installation entscheiden.

Datei-Explorer
2 Dateien

SKILL.md wird angezeigt

SKILL.md
Quellanweisungen · Schreibgeschützte Vorschau
name
your-skill-name
description
Use when <clear trigger condition for this skill>.
version
1.0.0
authors
["your-name-or-org"]
tags
["category","education"]
status
draft
# Purpose State the educational problem this skill solves and the intended outcome. # Activation Logic ## Activate When - Trigger 1 - Trigger 2 - Trigger 3 ## Do Not Activate When - Exclusion 1 - Exclusion 2 ## Priority Rules - If this skill conflicts with `<other-skill>`, prefer this skill when `<condition>`. # Inputs - Learner objective - Learner level signals - Repository or project context - Relevant prior conversation state # Workflow 1. **Orient** - Gather the minimum context required. - Confirm learner goal and level assumptions. 2. **Teach or Guide** - Deliver a concise step adapted to level. - Use one concrete example. 3. **Check** - Ask a reasoning-focused checkpoint question. 4. **Adapt** - Correct misconceptions or reduce complexity as needed. 5. **Advance** - Propose one next implementation action. # Educational Rules - Teach incrementally. - Require active learner reasoning. - Explain why a step matters. - Prefer practical implementation over passive summary. # Failure Modes and Guardrails 1. **Overload** - Symptom: long dense monologue. - Guardrail: limit to one core concept per step. 2. **Level Mismatch** - Symptom: learner confusion or overconfidence. - Guardrail: ask a checkpoint before increasing complexity. 3. **Answer Dumping** - Symptom: agent gives full solution immediately. - Guardrail: require learner attempt or prediction first. # Output Structure ```markdown ## Context <What we are solving and why> ## Guidance <Step-by-step explanation> ## Checkpoint <Question that forces reasoning> ## Next Step <Concrete implementation action> ``` # Validation Checklist - [ ] Activation triggers are explicit and testable. - [ ] Workflow includes at least one adaptation branch. - [ ] Educational rules are enforceable from output. - [ ] Failure modes include actionable guardrails. - [ ] Output structure is deterministic. - [ ] `examples.md` includes beginner, advanced, and recovery scenarios.
Auf GitHub ansehen