lazy-agent-loader
Load agent definitions on-demand to reduce context usage. Only loads full agent when needed.
14 stars
Best use case
lazy-agent-loader is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Load agent definitions on-demand to reduce context usage. Only loads full agent when needed.
Teams using lazy-agent-loader should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
$curl -o ~/.claude/skills/lazy-agent-loader/SKILL.md --create-dirs "https://raw.githubusercontent.com/nguyenthienthanh/aura-frog/main/aura-frog/skills/lazy-agent-loader/SKILL.md"
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/lazy-agent-loader/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How lazy-agent-loader Compares
| Feature / Agent | lazy-agent-loader | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Load agent definitions on-demand to reduce context usage. Only loads full agent when needed.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
> **AI-consumed reference.** Optimized for Claude to read during execution.
> Human-readable explanation: see [docs/architecture/HIERARCHICAL_PLANNING.md](../../../docs/architecture/HIERARCHICAL_PLANNING.md)
> or [docs/getting-started/](../../../docs/getting-started/) depending on topic.
# Lazy Agent Loader
Reduce token usage: load agent summaries initially (~50 tokens each), full definition only when activated.
---
## Agent Index
```toon
agent_index[16]{id,category,specialty,keywords}:
mobile,dev,React Native/Expo mobile,react-native/expo/RN/mobile/ios/android
mobile-flutter,dev,Flutter/Dart mobile,flutter/dart/bloc/mobile
web-angular,dev,Angular frontend,angular/ngrx/rxjs/typescript
web-vuejs,dev,Vue.js frontend,vue/vuejs/pinia/nuxt/composition
web-reactjs,dev,React frontend,react/reactjs/jsx/hooks/redux
web-nextjs,dev,Next.js fullstack,next/nextjs/ssr/ssg/app-router
architect,dev,Node.js backend,nodejs/express/nestjs/fastify/api
backend-python,dev,Python backend,python/django/fastapi/flask/api
backend-go,dev,Go backend,go/golang/gin/fiber/api
backend-laravel,dev,Laravel/PHP backend,laravel/php/eloquent/artisan
security,quality,Security auditing,security/vulnerability/audit/owasp
tester,quality,Testing/QA,test/testing/coverage/qa/jest/cypress
devops,ops,DevOps/CI-CD,deploy/docker/kubernetes/ci-cd/pipeline
router,infra,Agent detection,detect/agent/select/route
lead,infra,PM/orchestration,pm/project/orchestrate/manage
scanner,infra,Project management,project/detect/identify/config/context
```
---
## Loading Strategy
| Score | Level | Action |
|-------|-------|--------|
| >= 80 | PRIMARY | Load full definition (`agents/[id].md`) |
| 50-79 | SECONDARY | Summary only from index |
| < 50 | OPTIONAL | Don't load |
---
## Token Savings
```toon
comparison[4]{scenario,without_lazy,with_lazy,savings}:
Initial load,~48000,~1200,97.5%
Single agent,~48000,~2700,94.4%
Dual agent,~48000,~4200,91.3%
3 agents,~48000,~5700,88.1%
```
---
## Cache
Loaded agents cached in session. Track: `loaded_agents[]: mobile,tester`. Force reload: `reload agent <id>`.
---
## Integration
Used automatically by `agent-detector` for optimized loading. Score agents by keywords from index, load only PRIMARY agents.
---Related Skills
We are still matching the closest adjacent skills for this page. In the meantime, continue through the full directory.