菜单
Generate and deliver a learning exercise based on what the user is forgetting. Orchestrates different learning techniques.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Connect a notification channel (Slack, Gmail, etc.) so Cognition can send you learning exercises throughout the day — even when your laptop is off.
Schedule learning exercises for the next 24 hours via Slack/Gmail. Runs at the end of every session so exercises arrive even when Claude Code is closed. The API decides WHEN and WHAT technique — you build the exercise and schedule delivery.
Send screen observations to the Cognition API as learning events. Transforms Screenpipe data into API event format.
Launch the Cognition GUI app. Opens the Electron window and connects Claude Code to it via WebSocket. Use this after /cognition:start or whenever you need the GUI.
Test if the user's confidence matches their actual knowledge. Use when calibration ECE is high or strategy is "overconfident".
Quick surprise retention check. A 30-second mini-quiz to test if a concept is still in memory. Use for concepts with high stability that haven't been tested recently.
| name | learn |
| description | Generate and deliver a learning exercise based on what the user is forgetting. Orchestrates different learning techniques. |
This is the master learning skill. It decides which technique to use and executes it.
Get the data — call predict endpoints (recommendations, learner state, strategy signature, calibration)
Decide the technique based on the API response:
| Condition | Technique | Why |
|---|---|---|
| Lapses > 0 for this concept | Spaced Retrieval | They've forgotten before — need active recall |
| Recommendation action = "quiz" | Spaced Retrieval | API recommends quiz |
| Strategy = "overconfident" | Calibration Check | They think they know it but don't |
| Strategy = "cram_dependent" | Interleaving | Break the cramming pattern |
| Multiple related concepts decaying | Interleaving | Practice related concepts together |
| Concept has high stability but hasn't been tested | Delayed Probe | Quick surprise check |
| Recommendation action = "review" | Replay Consolidation | Multi-phase memory replay |
| User asks to explain something | Teach Back | Explain it to prove mastery |
curl -s -H "x-api-key: $TOKEN" \
"https://cognition-api.fly.dev/v1/stability/budget?user_id=USER_ID"
If the budget says the intervention would destabilize learning, defer or soften the exercise.
Execute the technique — load the specific technique skill and follow its instructions.
Record results via the reflect flow — POST quiz_answer or delayed_probe events back to the API.
Terminal mode (default): Present exercises as interactive markdown. Ask questions, wait for answers, give feedback.
Rich UI mode: When the exercise benefits from visual interaction (knowledge maps, multi-question quizzes), generate an HTML file using the scripts in scripts/ and open it in the browser:
python ${CLAUDE_SKILL_DIR}/../scripts/quiz.py --output /tmp/cognition-quiz.html
# then open it
Always show: