Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/primeline-ai/evolving-lite/system-bootnpx skills add primeline-ai/evolving-lite --skill system-bootgit clone --depth 1 https://github.com/primeline-ai/evolving-liteWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00039 | $0.00600 |
| Opus 5 | $0.00019 | $0.00300 |
| Sonnet 5 | $0.00008 | $0.00120 |
| Haiku 4.5 | $0.00004 | $0.00060 |
Grade A, and why
system-boot scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
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 onlast_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 entriesnext_step: What was planned nextfailures: 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":
- Read the most recent session summary from
${CLAUDE_PLUGIN_ROOT}/_memory/sessions/ - Load the plan if one was referenced
- Continue immediately - no questions, just pick up where we left off
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 70 lines · 39 tokens per session scan A 957dc209c093
system-boot is a skill published in the GitHub repository primeline-ai/evolving-lite (48 stars, last pushed 16d ago), licensed MIT. It adds 39 tokens to every session and 600 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
subagent-strategy
Delegates research and parallel work to sub-agents.
project-brain
Triggers (all require explicit user request — do NOT activate just because a brain/ folder exists): (1) The user wants to set up project-level memory ("set up project brain", "scaffold project context", "init project brain", "建项目脑"). (2) The user is in a directory containing brain/ AND explicitly asks to resume /…
dispatching-parallel-agents
Runs independent subtasks concurrently via sub-agents.
durable-work
Picks the board over sub-agents for cross-turn work.
spawn-agent
Spawn worker agents (Gemini CLI or Codex CLI) to keep main context clean. Use for implementation, codebase research, context gathering, or any scoped work that would pollute the orchestrator's context.
brand-guidelines
Platform-agnostic brand identity creation using 12 Jungian archetypes. Use when running /mybrand, creating brand.md, selecting color palettes, defining typography, or asking about brand personality and voice guidelines.