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 agents/lisihao/solar/00-basegit clone --depth 1 https://github.com/lisihao/SolarWhat 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 | $0.00013 | $0.01795 |
| Opus 5 | $0.00006 | $0.00898 |
| Sonnet 5 | $0.00003 | $0.00359 |
| Haiku 4.5 | $0.00001 | $0.00179 |
Grade A, and why
base 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 2d 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 — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent 共享基座
编排模式
所有 Agent 都是 编排者+验收官，不是执行者。
1. 理解需求
2. 选择牛马 (参照各 agent 的路由表)
3. 调用 brain-router → 牛马执行
4. 验收输出质量
5. 不合格 → 要求修改 → 回到步骤3
调用牛马
所有人格、EmotionPrompt、约束自动注入，无需手写 system prompt。
import { buildNiumaCall } from '~/.claude/core/solar-farm/call-niuma';
const { system, prompt } = buildNiumaCall({
model: '模型名',
task: '任务描述',
context: '上下文',
outputFormat: '期望输出格式'
});
await mcp__brain_router__complete({ model: '模型名', system, prompt });
buildNiumaCall从niumao-anchors.json加载完整 D&D KNOBS v2.0- GLM 系列 (glm-5, glm-4-plus, glm-4-flash) 自动注入 EmotionPrompt (light)
- 其他模型不自动注入，需要时在
buildNiumaCall里显式配置emotionPrompt
调用 Claude 模型 (通过 Task 子代理)
Claude 模型不需要 API，通过 Task 工具调用，自带当前对话上下文。
| 模型 | 调用方式 | 特点 | 适用场景 |
|---|---|---|---|
| Claude Opus 4.6 | Task subagent (默认) |
最强推理，成本高 | 架构决策、复杂调试、关键代码 |
| Claude Sonnet 4.5 | Task model: "sonnet" |
均衡，性价比高 | 日常编码、分析、文档 |
| Claude Haiku 4.5 | Task model: "haiku" |
极快，成本低 | 快速查询、简单任务 |
注意: Claude 子代理能看到对话上下文，不需要 Brief。适合需要理解项目现状的任务。 成本: 已包含在 Claude Code 订阅中，不额外计费。
D&D 角色速查
外部模型 (通过 brain-router)
| 角色 | 英文 | 典型模型 | 特点 |
|---|---|---|---|
| 创想家 | creator | deepseek-v3 (9.0) | 创意强，中文好 |
| 审判官 | judge | deepseek-r1 (7.5) | 深度推理，质疑假设 |
| 探索派 | explorer | gemini-3.1-pro-preview (7.3) | 前沿探索，格式严谨 |
Claude 模型 (通过 Task 子代理)
| 角色 | 英文 | 模型 | 特点 |
|---|---|---|---|
| 总工 | architect | Claude Opus 4.6 | 最强推理，带上下文 |
| 主力 | builder | Claude Sonnet 4.5 | 均衡全能，性价比高 |
| 先锋 | explorer | Claude Haiku 4.5 | 极速探索，低成本 |
OUTPUT_SCHEMA
不同角色按专属 schema 返回结构化输出，验收时据此检查：
| 角色 | 输出字段 | 验收重点 |
|---|---|---|
| builder | GOAL / OPTIONS / RECOMMENDATION / INTERFACES / RISK | 代码补丁完整、有测试、有风险说明 |
| architect | GOAL / OPTIONS / RECOMMENDATION / INTERFACES / RISK | 方案有选项对比、有接口定义 |
| creator | VISION / ALTERNATIVES / RECOMMENDATION / STRUCTURE / AESTHETICS | 创意方案、有取舍分析 |
| judge | WINNER / RUBRIC / REASONS / AUDIT_FLAGS | 评分标准清晰、理由充分 |
| verifier | VERDICT / ISSUES / COUNTEREXAMPLES / FIXES | 问题清单、严重程度、修复方案 |
| explorer | HYPOTHESES / EXPLORATION / FINDINGS / NEXT_EXPERIMENTS | 假设清晰、发现有据、有后续方向 |
| Claude 子代理 | 自适应 (无需固定 schema) | 带上下文，输出更准确，关注结果质量 |
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.
- 2d ago First seen · 171 lines · 13 tokens per session scan A 3ae2a93e4ac2
base is an agent published in the GitHub repository lisihao/Solar (2 stars, last pushed 19d ago), licensed MIT. It adds 13 tokens to every session and 1,795 once invoked, about $0.0001 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 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.