スキル一覧に戻る

basic-obedience

pjt222
更新日 2 days ago
7 閲覧
17
2
17
GitHubで表示
その他ai

について

このスキルは、陽性強化を用いて「お座り」「待て」「おいで」などの基本コマンドを教え、犬の基礎服従訓練を提供します。タイミングの取り方、セッションの組み立て方、気が散る環境での確実性向上など、重要な訓練要素を網羅しています。新しい子犬、確実な基礎を必要とする成犬、保護犬、高度な訓練前の劣化したコマンドの再学習にご活用ください。

クイックインストール

Claude Code

推奨
メイン
npx skills add pjt222/agent-almanac -a claude-code
プラグインコマンド代替
/plugin add https://github.com/pjt222/agent-almanac
Git クローン代替
git clone https://github.com/pjt222/agent-almanac.git ~/.claude/skills/basic-obedience

このコマンドをClaude Codeにコピー&ペーストしてスキルをインストールします

ドキュメント

Grundgehorsam

Lehren foundation commands (sit, stay, come, heel, down) using positive reinforcement and marker training.

Wann verwenden

  • A new puppy (8+ weeks) is ready for foundation training
  • An adult dog lacks reliable basic commands
  • A rescue or rehomed dog needs to learn the household's command vocabulary
  • Before advancing to more complex behaviors or off-leash work
  • When existing commands have degraded and need re-establishing

Eingaben

  • Erforderlich: A dog (any breed, any age 8+ weeks)
  • Erforderlich: High-value treats (small, soft, quickly consumed)
  • Optional: Clicker or verbal marker word (e.g., "yes")
  • Optional: 6-foot leash and flat collar or harness
  • Optional: Quiet training space with minimal distractions (initially)

Vorgehensweise

Schritt 1: Establish the Marker

The marker bridges the gap zwischen the desired behavior and the reward.

Marker Training Protocol:
1. Choose your marker: clicker (precise) or verbal "yes" (always available)
2. Charge the marker (10-15 reps):
   - Mark (click or "yes") then immediately deliver a treat
   - No behavior required — just marker → treat, marker → treat
   - Dog should begin orienting toward you at the sound of the marker
3. Test: mark when the dog is looking away. Does the dog turn toward
   you expecting a treat? If yes, the marker is charged.

Timing Rule:
The marker must occur WITHIN 1 second of the desired behavior.
Late marking teaches the wrong behavior.
Mark → then reach for the treat (not the reverse).

Erwartet: The dog reliably orients toward the handler upon hearing the marker, expecting a reward.

Bei Fehler: If the dog nicht respond to the marker nach 20 reps, the treat value is too low. Switch to higher-value rewards (cheese, chicken, liver). If the dog is too distracted to eat, die Umgebung is too stimulating — move to a quieter space.

Schritt 2: Lehren the Five Foundation Commands

Work on one command per session until reliable, then begin mixing.

Command Protocols:

SIT:
1. Hold treat above dog's nose, slowly arc backward over the head
2. As the dog's head follows up, the rear naturally lowers
3. The instant the rear touches the ground → mark and treat
4. Add the verbal cue "sit" AFTER the dog is offering the behavior reliably
   (cue comes before behavior only once the dog understands the behavior)

DOWN:
1. From a sit, hold treat at the dog's nose then lower slowly to the ground
2. Draw the treat slightly forward along the ground
3. As elbows touch the ground → mark and treat
4. If the dog stands instead, reset and try with less forward movement

STAY:
1. Ask for a sit or down
2. Open palm toward the dog, say "stay"
3. Wait 1 second → mark and treat while the dog is still in position
4. Gradually increase duration: 2s, 5s, 10s, 30s, 1 min
5. Add distance: one step back, then two, then five
6. Add distraction: only after duration and distance are solid
   (the "three Ds": Duration, Distance, Distraction — increase one at a time)

COME (recall):
1. Start on a long line (15-30 ft) in a low-distraction environment
2. Let the dog wander, then call name + "come" in an upbeat tone
3. If the dog turns toward you → mark → reward generously when the dog arrives
4. NEVER call "come" for something unpleasant (bath, crate, leaving the park)
5. Recall is the most important safety command — make it the most rewarding

HEEL:
1. Dog on your left side, treat in left hand at your hip
2. Take one step, if the dog moves with you → mark and treat
3. Gradually increase to two steps, five steps, ten steps
4. Mark and treat for maintaining position (head roughly at your knee)
5. If the dog pulls ahead, stop walking. Resume when the leash is loose.

Erwartet: Each command is performed reliably in a low-distraction environment with treats as motivation.

Bei Fehler: If a command ist nicht progressing nach 3 sessions, break it into smaller steps. The dog may need an intermediate behavior (e.g., for "down," reward the head-lowering motion vor requiring full elbows-on-ground).

Schritt 3: Structure Training Sessions

Session Guidelines:
+--------------------+------------------------------------------+
| Parameter          | Guideline                                |
+--------------------+------------------------------------------+
| Duration           | 5-10 minutes (puppies: 3-5 minutes)      |
| Frequency          | 2-3 sessions per day                     |
| End on success     | Always end after a successful rep, not   |
|                    | after a failure                          |
| Reward rate        | Initially: every correct rep             |
|                    | Later: intermittent (variable schedule)  |
| Energy management  | High-energy dog? Exercise BEFORE training|
|                    | Low-energy dog? Train when most alert    |
| Session structure  | Warm-up (easy known command) → new       |
|                    | material → cool-down (easy command)      |
+--------------------+------------------------------------------+

The 80/20 Rule:
- 80% of reps should succeed (dog is getting it right)
- If success rate drops below 80%, the criteria is too high — go easier
- 20% challenge keeps the dog engaged without frustrating

Erwartet: Short, successful sessions that leave the dog wanting more.

Bei Fehler: If the dog disengages (sniffing, looking away, lying down), the session is too long, too hard, or the rewards are insufficiently motivating. End the session and reassess.

Schritt 4: Distraction-Proof the Commands

Once reliable in a quiet environment, systematically add distractions.

Distraction Ladder (work through sequentially):
1. Quiet room, no distractions (starting point)
2. Room with a family member present
3. Backyard or garden
4. Front yard with street noise
5. Quiet park or trail
6. Busy park with other dogs at a distance
7. Busy park with other dogs nearby
8. Novel environments (pet store, cafe patio)

At each new level:
- Expect performance to decrease — this is normal
- Increase reward rate back to every correct rep
- Do not add more distraction until the current level is reliable
- If the dog fails 3 reps in a row, you moved up too fast — go back one level

Erwartet: Commands work reliably in progressively more distracting environments.

Bei Fehler: If a specific distraction (other dogs, squirrels) consistently breaks training, that distraction needs separate counter-conditioning work (see behavioral-modification).

Validierung

  • Marker is charged and the dog responds reliably
  • All five commands are performed in a low-distraction environment
  • Training sessions are 5-10 minutes, ending on success
  • Success rate is at or ueber 80% fuer jede command
  • Commands are being generalized durch the distraction ladder
  • The handler's timing (marker innerhalb 1 second) is consistent

Haeufige Stolperfallen

  • Repeating the cue: Saying "sit, sit, SIT" teaches the dog that the first "sit" is optional. Say it once and wait
  • Treating too late: The treat should follow the marker innerhalb 2-3 seconds. Late treating breaks the association
  • Luring forever: The hand motion with the treat (lure) sollte faded innerhalb 10-20 reps. Otherwise the dog only responds when food is visible
  • Punishing failed recalls: Calling "come" and then scolding the dog (for being slow, for having something in its mouth) poisons the recall cue permanently
  • Training too long: A fatigued dog learns nothing. Quit while ahead
  • Inconsistent cues: All household members must use the same words and gestures fuer jede command

Verwandte Skills

  • behavioral-modification — for addressing unwanted behaviors that interfere with basic obedience

GitHub リポジトリ

pjt222/agent-almanac
パス: i18n/de/skills/basic-obedience
0
agentsagentskillsai-assisted-developmentclaude-codeskillsteams

関連スキル

llamaguard

その他

LlamaGuardは、暴力やヘイトスピーチなど6つの安全性カテゴリーにおいて、LLMの入力と出力をモデレートするMetaの70-80億パラメータモデルです。94〜95%の精度を提供し、vLLM、Hugging Face、Amazon SageMakerを使用してデプロイ可能です。このスキルを使用して、AIアプリケーションにコンテンツフィルタリングと安全策を簡単に統合できます。

スキルを見る

cost-optimization

その他

このClaudeスキルは、リソースの適正サイジング、タグ付け戦略、支出分析を通じて、開発者がクラウドコストを最適化することを支援します。AWS、Azure、GCPにわたるクラウド支出の削減とコストガバナンスの実施のためのフレームワークを提供します。インフラコストの分析、リソースの適正サイジング、または予算制約への対応が必要な際にご利用ください。

スキルを見る

quantizing-models-bitsandbytes

その他

このスキルは、bitsandbytesを使用してLLMを8ビットまたは4ビット精度に量子化し、精度の低下を最小限に抑えつつ50〜75%のメモリ削減を実現します。限られたGPUメモリでより大規模なモデルを実行したり、推論を高速化するのに理想的で、INT8、NF4、FP4などのフォーマットをサポートしています。HuggingFace Transformersと統合され、QLoRAトレーニングや8ビットオプティマイザーを可能にします。

スキルを見る

dispatching-parallel-agents

その他

このClaudeスキルは、複数のエージェントを配備し、3つ以上の独立した問題を並行して調査・修正します。共有状態や依存関係がなく解決可能な、無関係な障害が発生するシナリオ向けに設計されています。中核となる機能は並列問題解決であり、効率を最大化するために独立した問題領域ごとに1つのエージェントを割り当てます。

スキルを見る