From d-nd-core
Detect gaps, implement fixes, crystallize learnings. Use when something fails, when a pattern repeats, or when the system encounters a problem it should not encounter again.
How this skill is triggered — by the user, by Claude, or both
Slash command
/d-nd-core:auto-learnThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
When something goes wrong, this skill ensures it never goes wrong the same way again.
When something goes wrong, this skill ensures it never goes wrong the same way again.
Something failed, broke, or required manual correction:
Ask: "What went wrong? What was the gap?"
Identify the root cause — not the symptom:
Ask: "Why did the system allow this to happen?"
Fix the gap structurally — not with a workaround:
One fix per gap. Do not batch. Each fix is complete on its own.
Record the learning so it persists:
Create a file in .claude/memory/ with this structure:
---
name: [what was learned]
description: [one line — when this applies]
type: feedback
---
[The rule or pattern]
**Why:** [what happened that led to this]
**How to apply:** [when and where this kicks in]
Test that the fix works:
Ask: "Who else in the system needs this?"
This skill activates automatically when:
You do not need to be told to learn. You need to learn when it happens.
The system that corrects itself once does not make the same mistake twice. The correction lives in the seed. The seed propagates. Every instance that comes after inherits the learning without having made the error.
This is how the system sustains itself. Not by being perfect — by making perfection unnecessary.
npx claudepluginhub grazianoguiducci/d-nd-seed --plugin d-nd-coreLogs user corrections like 'wrong' or 'no' as mistakes, updates CLAUDE.md rules, and auto-generates skills from repeated patterns after 3+ occurrences. Triggers on feedback or repetitions.
Captures agent mistakes, corrections, and discovered gotchas so they are not repeated. Use when: (1) a command or operation fails unexpectedly, (2) the user corrects the agent, (3) the agent discovers non-obvious behavior through debugging, (4) an API or tool behaves differently than expected, (5) a better approach is found for a recurring task. Also searches past learnings before starting tasks to avoid known pitfalls. Activate alongside the memory skill — they share sage-memory but serve different purposes (memory = codebase knowledge, self-learning = agent mistakes and gotchas). Also trigger on "sage review" or "review learnings" to curate and improve the learning database.
Logs errors, user corrections, missing features, API failures, knowledge gaps, and best practices to .learnings/ markdown files. Promotes key insights to CLAUDE.md and AGENTS.md for AI agent self-improvement.