Receber por email
← Voltar para o catálogo

lessons-consolidate

Periodic maintenance of institutional knowledge. Analyzes lessons for staleness, overlaps, and gaps. Keeps lessons.json healthy.

17estrelas
Atualizado há 5 meses

Ver no GitHub ↗ Licença: MIT

Como adicionar

/plugin marketplace add walm00/business-context-os

O comando exato pode variar conforme o repositório. Confira o README no GitHub.

Para o autor da skill

Cole no README do seu repo

Mostra que sua skill está catalogada na Skillteca, gera backlink e tráfego rastreável.

[![Skillteca](https://www.skillteca.com.br/api/badge/lessons-consolidate/svg)](https://www.skillteca.com.br/skills/lessons-consolidate)
#ai

Alerta por categoria

Receba novas skills de Escrita e Conteúdo toda segunda

1 email curto, só com as skills novas de Escrita e Conteúdo. 4 minutos de leitura, sem spam, cancela com 1 clique.

Você confirma o email no primeiro envio. Sem spam. Cancela com 1 clique.

Ler descrição completa

Lessons Consolidate

Purpose

This skill IS:

  • A periodic curator of institutional knowledge in lessons.json
  • A tool for detecting stale, overlapping, or contradictory lessons
  • A way to keep the lessons collection healthy and useful

This skill IS NOT:

  • For capturing new lessons (use ecosystem-manager for that)
  • Automatically scheduled (invoke it periodically)
  • A replacement for reading lessons (it maintains, doesn't apply them)

When to Use

  • Monthly or every 10-15 sessions (whichever comes first)
  • When lessons.json feels cluttered
  • When you suspect lessons are outdated
  • When you want to understand patterns across sessions

Modes

1. Analyze

Check the current state of lessons.json:

python .claude/scripts/consolidate_lessons.py

Identifies:

  • Stale lessons: Referenced concepts that no longer exist
  • Overlapping lessons: Multiple lessons saying the same thing
  • Contradictions: Lessons that conflict with each other
  • Tag gaps: Important topics with no lessons
  • Usage patterns: Which tags are over/under-represented

2. Consolidate

After analysis, with user approval:

  • Archive stale lessons (don't delete - move to archived section)
  • Merge overlapping lessons into single stronger lessons
  • Flag contradictions for user resolution
  • Suggest new lessons for uncovered areas

3. Report

Generate a health summary:

  • Total lessons and trend
  • Tag distribution
  • Staleness assessment
  • Recommendations for next actions

Guidelines

Good lessons are:

  • Specific enough to act on
  • General enough to apply to future situations
  • Tagged with relevant categories
  • Tied to a real experience (not hypothetical)

Boundary: Lessons vs Wiki

Use this boundary when bcos-wiki exists:

| Content | Owning surface | |---------|----------------| | Atomic, reusable rule for future agents or humans | lessons.json via ecosystem-manager | | Narrative incident explanation, timeline, causes, and follow-up context | Wiki post-mortem page-type | | Decision narrative, alternatives considered, and why a choice was made | Wiki decision-log page-type | | Step-by-step operational response | Wiki how-to or runbook page-type |

Lessons should stay short and actionable. Wiki pages can carry the richer story. If consolidation finds a lesson that is really a narrative, recommend moving the detail to /wiki create and keeping only the distilled lesson in lessons.json.

Cross-reference: docs/_bcos-framework/architecture/wiki-zone.md defines wiki page-type semantics and schema governance.

Remove or archive when:

  • The lesson references something that no longer exists
  • The lesson has been superseded by a better one
  • The lesson is too vague to be useful
  • The lesson contradicts current practice

References

  • Lessons file: .claude/quality/ecosystem/lessons.json
  • Schema: .claude/quality/ecosystem/lessons-schema.md
  • Consolidation script: .claude/scripts/consolidate_lessons.py
Compartilhar X LinkedIn