Receber por email
← Voltar para o catálogo

system-boot

Auto-activating session startup - reads memory, orients Claude, announces status. Triggers on session start or when user says "continue", "weiter", "resume".

48estrelas
Atualizado há 4 meses

Ver no GitHub ↗ Licença: MIT

Como adicionar

/plugin marketplace add primeline-ai/evolving-lite

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/system-boot/svg)](https://www.skillteca.com.br/skills/system-boot)

Skills relacionadas

Ver todas de Marketing →

Alerta por categoria

Receba novas skills de Marketing toda segunda

1 email curto, só com as skills novas de Marketing. 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

System Boot

On every session start, perform this bootup sequence:

1. Read Memory

Read ${CLAUDE_PLUGIN_ROOT}/_memory/index.json for:

  • active_project: Which project the user was working on
  • last_session: When the last session was

If active_project is set, read ${CLAUDE_PLUGIN_ROOT}/_memory/projects/{active_project}.json for:

  • progress: Array of recent progress entries
  • next_step: What was planned next
  • failures: Known blockers or failures

2. Orient

Based on memory:

  • What was the last progress entry?
  • Are there any known failures or blockers?
  • What was suggested as the next step?

3. Announce

Output a compact status line:

Evolving Lite | Session {n} | Tier {tier} | {experience_count} experiences
Last: {last_progress_summary}
Next: {next_step}

If this is the first session ever (no progress), announce with a short, visible cold-start sequence so the system does not feel inert on day 1:

Evolving Lite v1.0 | First session
  [1/3] Loading {prewarmed_count} pre-warmed experiences ... ok
  [2/3] Tier 1 (Safety) active: context warnings, bash security, hook sentinels
  [3/3] Self-Star Doctor: wiring verified (run /health for the full board)
Ready. The system learns from your corrections automatically and self-tunes its
delegation routing from your sessions (see "Self-Evolution is ON" in the README).

The Self-Star Doctor also runs a quick wiring + preflight check automatically on this first session. If it reports a missing prerequisite (most often Kairn), surface that to the user and point them at /health and pip install kairn-ai.

4. Pick One Task

If the user hasn't specified what to work on, suggest the next_step from memory. Don't start multiple tasks - pick one.

Continue Trigger

When the user says "continue", "weiter", "weitermachen", "fortsetzen", or "resume":

  1. Read the most recent session summary from ${CLAUDE_PLUGIN_ROOT}/_memory/sessions/
  2. Load the plan if one was referenced
  3. Continue immediately - no questions, just pick up where we left off

Session End

When a session ends naturally (user says goodbye, or Stop hook fires):

  • The session-summary hook handles progress logging automatically
  • No manual action needed from the user
Compartilhar X LinkedIn