Skip to main content

ooda-loop

Outmaneuver competition by cycling through observe-orient-decide-act faster than opponents, operating inside their decision-making tempo

ソース情報

リポジトリ
lev-os/agents
ソースの最終更新活動
2026年3月7日 00:14
検出された SKILL.md の言語
英語
スター
22
フォーク
2

インストール方法

デフォルトでは、最初にソースを確認する Prompt が選択されています。直接コマンドに切り替えるか、ローカルコピーをダウンロードすることもできます。

ソースファイルを確認

インストールを決める前に、SKILL.md と SkillsMP に表示されている付属ファイルをお読みください。

SKILL.md を表示中

SKILL.md
ソースの指示 · 読み取り専用プレビュー
name
ooda-loop
description
Outmaneuver competition by cycling through observe-orient-decide-act faster than opponents, operating inside their decision-making tempo
# OODA Loop (Boyd Cycle) ## Overview The OODA Loop (Observe, Orient, Decide, Act) is a decision-making framework developed by U.S. Air Force Colonel John Boyd in the early 1970s, based on his experiences as a Korean War fighter pilot. Boyd discovered that victory in competitive environments doesn't come from having better resources but from cycling through decisions faster than your opponent - "getting inside their OODA loop." The framework is deceptively simple: observe the situation, orient to understand context, decide on a course of action, and act. Then repeat continuously. The critical insight is tempo: the entity that can process this cycle more rapidly and effectively gains a decisive advantage. Boyd emphasized that "Orient" is the most crucial step - it's where you contextualize observations against mental models, past experiences, and cultural context to make the right kind of decision. ## When to Use - Competitive environments requiring rapid adaptation (business, sports, conflict) - Situations changing faster than traditional planning cycles can handle - You need to disrupt an opponent's rhythm or decision-making process - Dynamic markets where early movers capture advantage - Crisis response requiring continuous situational reassessment - Building organizational agility and adaptive capacity - Diagnosing why you're losing to faster-moving competitors ## The Process ### Step 1: Observe - Gather Raw Information Collect unfiltered data about the current situation from all available sources. Don't interpret yet. **Example:** Competitor launched new feature, user churn increased 15%, support tickets spiked, social media sentiment turned negative. ### Step 2: Orient - Contextualize and Interpret This is the decisive step. Filter observations through your mental models, cultural context, past experiences, and expertise to understand what the data means. **Critical elements:** Genetic heritage, cultural traditions, previous experiences, unfolding circumstances, new information analysis, synthesis, and feedback. **Example:** Orient the observations - competitor's feature targets our core weakness, churn is price-sensitive segment, tickets show confusion not bugs, sentiment driven by influencers we don't reach. ### Step 3: Decide - Choose Course of Action Based on your oriented understanding, select the best response from available options. **Example:** Decide to fast-track competing feature development (3 weeks vs. 3 months), launch retention campaign for price-sensitive segment, engage influencers directly. ### Step 4: Act - Execute the Decision Implement your decision. Execution creates new observations, feeding the next cycle. **Example:** Act - reassign engineering team, launch discount offer, reach out to 10 influencers. These actions change the competitive landscape. ### Step 5: Cycle Continuously - Tempo is Everything The loop never stops. As you act, immediately begin observing the results and opponents' reactions. Faster cycling creates disorientation in slower opponents. **Tempo advantage:** If your cycle takes 1 week and competitors take 4 weeks, you complete 4 iterations while they complete 1 - they're always reacting to stale information. ## Example Application **Situation (Boyd's fighter pilot insight):** U.S. F-86 fighters had 10:1 kill ratio against superior Soviet MiG-15s in Korean War despite MiG's better speed and climb rate. **Application:** F-86 pilots could observe enemy maneuvers, orient faster (better cockpit visibility, hydraulic controls), decide on counter-maneuvers, and act (execute turns) more rapidly than MiG pilots. This faster OODA cycle let them dictate engagement terms. **Outcome:** Pilots who completed 3-4 OODA cycles during a dogfight could position inside the enemy's decision loop - by the time the MiG pilot decided on a maneuver based on F-86's position, the F-86 had already moved twice, making the MiG's decision obsolete. ## Example Application 2 **Situation:** Startup competing against established enterprise software company. **Application:** - **Observe:** Enterprise ships quarterly; we ship weekly. Customer needs evolve monthly. - **Orient:** Their slow cycle means they build to 4-month-old requirements; ours are 1 week old. - **Decide:** Optimize for iteration speed over feature completeness. - **Act:** Ship MVP features, gather feedback, iterate. Each week we complete full OODA cycle. **Outcome:** After 3 months, startup has iterated 12 times vs. enterprise's 1 iteration. Product fits current market; enterprise's doesn't. Startup wins key accounts despite fewer features. ## Anti-Patterns - ❌ Treating OODA as linear sequence instead of continuous cycle - ❌ Obsessing over speed at Orient step - thorough orientation is more important than fast decisions - ❌ Neglecting to observe results of your actions (breaking the feedback loop) - ❌ Using stale mental models in Orient step (past experience becomes liability) - ❌ Focusing only on your OODA loop without disrupting opponent's loop - ❌ Optimizing individual steps instead of overall cycle time - ❌ Forgetting Boyd's insight: Orient is the decisive function, not Decide ## Related - second-order-thinking (anticipating opponent's reactions to your actions) - tempo-advantage (speed as competitive weapon) - agile-methodology (software development's continuous iteration) - feedback-loops (observing results to inform next cycle) - fog-of-war (incomplete information in competitive environments)
GitHubで見る