docs: update AGENTS.md hierarchy with latest structure and line counts
- Root: Add Prometheus/Metis/Momus agents, MCP architecture, 82 test files - agents/: Document 7-section delegation and wisdom notepad - auth/: Multi-account load balancing (10 accounts), endpoint fallback - features/: Update background-agent 825 lines, builtin-skills 1230 lines - hooks/: 22+ hooks with event timing details - tools/: sisyphus-task 583 lines, LSP client 632 lines - cli/: config-manager 725 lines, 17+ doctor checks - shared/: Cross-cutting utilities with usage patterns
This commit is contained in:
YeonGyu-Kim
+32
-68
@@ -1,25 +1,23 @@
|
||||
# AGENTS KNOWLEDGE BASE
|
||||
|
||||
## OVERVIEW
|
||||
|
||||
AI agent definitions for multi-model orchestration. 7 specialized agents: Sisyphus (orchestrator), oracle (read-only consultation), librarian (research), explore (grep), frontend-ui-ux-engineer, document-writer, multimodal-looker.
|
||||
AI agent definitions for multi-model orchestration, delegating tasks to specialized experts.
|
||||
|
||||
## STRUCTURE
|
||||
|
||||
```
|
||||
agents/
|
||||
├── orchestrator-sisyphus.ts # Orchestrator agent (1484 lines) - complex delegation
|
||||
├── sisyphus.ts # Main Sisyphus prompt (641 lines)
|
||||
├── orchestrator-sisyphus.ts # Orchestrator agent (1486 lines) - 7-section delegation, wisdom
|
||||
├── sisyphus.ts # Main Sisyphus prompt (643 lines)
|
||||
├── sisyphus-junior.ts # Junior variant for delegated tasks
|
||||
├── oracle.ts # Strategic advisor (GPT-5.2)
|
||||
├── librarian.ts # Multi-repo research (Claude Sonnet 4.5)
|
||||
├── librarian.ts # Multi-repo research (GLM-4.7-free)
|
||||
├── explore.ts # Fast codebase grep (Grok Code)
|
||||
├── frontend-ui-ux-engineer.ts # UI generation (Gemini 3 Pro)
|
||||
├── document-writer.ts # Technical docs (Gemini 3 Pro)
|
||||
├── multimodal-looker.ts # PDF/image analysis (Gemini 3 Flash)
|
||||
├── prometheus-prompt.ts # Planning agent prompt (982 lines)
|
||||
├── metis.ts # Plan Consultant agent (404 lines)
|
||||
├── momus.ts # Plan Reviewer agent (404 lines)
|
||||
├── prometheus-prompt.ts # Planning agent prompt (988 lines) - interview mode
|
||||
├── metis.ts # Plan Consultant agent - pre-planning analysis
|
||||
├── momus.ts # Plan Reviewer agent - plan validation
|
||||
├── build-prompt.ts # Shared build agent prompt
|
||||
├── plan-prompt.ts # Shared plan agent prompt
|
||||
├── types.ts # AgentModelConfig interface
|
||||
@@ -28,69 +26,35 @@ agents/
|
||||
```
|
||||
|
||||
## AGENT MODELS
|
||||
|
||||
| Agent | Default Model | Fallback | Purpose |
|
||||
|-------|---------------|----------|---------|
|
||||
| Sisyphus | anthropic/claude-opus-4-5 | - | Primary orchestrator with extended thinking |
|
||||
| oracle | openai/gpt-5.2 | - | Read-only consultation. High-IQ debugging, architecture |
|
||||
| librarian | opencode/glm-4.7-free | - | Docs, OSS research, GitHub examples |
|
||||
| explore | opencode/grok-code | google/gemini-3-flash, anthropic/claude-haiku-4-5 | Fast contextual grep |
|
||||
| frontend-ui-ux-engineer | google/gemini-3-pro-preview | - | UI/UX code generation |
|
||||
| document-writer | google/gemini-3-pro-preview | - | Technical writing |
|
||||
| multimodal-looker | google/gemini-3-flash | - | PDF/image analysis |
|
||||
| Agent | Default Model | Purpose |
|
||||
|-------|---------------|---------|
|
||||
| Sisyphus | claude-opus-4-5 | Primary orchestrator. 32k extended thinking budget. |
|
||||
| oracle | openai/gpt-5.2 | High-IQ debugging, architecture, strategic consultation. |
|
||||
| librarian | glm-4.7-free | Multi-repo analysis, docs research, GitHub examples. |
|
||||
| explore | grok-code | Fast contextual grep. Fallbacks: Gemini-3-Flash, Haiku-4-5. |
|
||||
| frontend-ui-ux | gemini-3-pro | Production-grade UI/UX generation and styling. |
|
||||
| document-writer | gemini-3-pro | Technical writing, guides, API documentation. |
|
||||
| Prometheus | claude-opus-4-5 | Strategic planner. Interview mode, orchestrates Metis/Momus. |
|
||||
| Metis | claude-sonnet-4-5 | Plan Consultant. Pre-planning risk/requirement analysis. |
|
||||
| Momus | claude-sonnet-4-5 | Plan Reviewer. Validation and quality enforcement. |
|
||||
|
||||
## HOW TO ADD AN AGENT
|
||||
|
||||
1. Create `src/agents/my-agent.ts`:
|
||||
```typescript
|
||||
import type { AgentConfig } from "@opencode-ai/sdk"
|
||||
|
||||
export const myAgent: AgentConfig = {
|
||||
model: "provider/model-name",
|
||||
temperature: 0.1,
|
||||
system: "Agent system prompt...",
|
||||
tools: { include: ["tool1", "tool2"] }, // or exclude: [...]
|
||||
}
|
||||
```
|
||||
2. Add to `builtinAgents` in `src/agents/index.ts`
|
||||
3. Update `types.ts` if adding new config options
|
||||
|
||||
## AGENT CONFIG OPTIONS
|
||||
|
||||
| Option | Type | Description |
|
||||
|--------|------|-------------|
|
||||
| model | string | Model identifier (provider/model-name) |
|
||||
| temperature | number | 0.0-1.0, most use 0.1 for consistency |
|
||||
| system | string | System prompt (can be multiline template literal) |
|
||||
| tools | object | `{ include: [...] }` or `{ exclude: [...] }` |
|
||||
| top_p | number | Optional nucleus sampling |
|
||||
| maxTokens | number | Optional max output tokens |
|
||||
1. Create `src/agents/my-agent.ts` exporting `AgentConfig`.
|
||||
2. Add to `builtinAgents` in `src/agents/index.ts`.
|
||||
3. Update `types.ts` if adding new config interfaces.
|
||||
|
||||
## MODEL FALLBACK LOGIC
|
||||
`createBuiltinAgents()` handles resolution:
|
||||
1. User config override (`agents.{name}.model`).
|
||||
2. Environment-specific settings (max20, antigravity).
|
||||
3. Hardcoded defaults in `index.ts`.
|
||||
|
||||
`createBuiltinAgents()` in utils.ts handles model fallback:
|
||||
|
||||
1. Check user config override (`agents.{name}.model`)
|
||||
2. Check installer settings (claude max20, gemini antigravity)
|
||||
3. Use default model
|
||||
|
||||
**Fallback order for explore**:
|
||||
- If gemini antigravity enabled → `google/gemini-3-flash`
|
||||
- If claude max20 enabled → `anthropic/claude-haiku-4-5`
|
||||
- Default → `opencode/grok-code` (free)
|
||||
|
||||
## ANTI-PATTERNS (AGENTS)
|
||||
|
||||
- **High temperature**: Don't use >0.3 for code-related agents
|
||||
- **Broad tool access**: Prefer explicit `include` over unrestricted access
|
||||
- **Monolithic prompts**: Keep prompts focused; delegate to specialized agents
|
||||
- **Missing fallbacks**: Consider free/cheap fallbacks for rate-limited models
|
||||
## ANTI-PATTERNS
|
||||
- **Trusting reports**: NEVER trust subagent self-reports; always verify outputs.
|
||||
- **High temp**: Don't use >0.3 for code agents (Sisyphus/Prometheus use 0.1).
|
||||
- **Sequential calls**: Prefer `sisyphus_task` with `run_in_background` for parallelism.
|
||||
|
||||
## SHARED PROMPTS
|
||||
|
||||
- **build-prompt.ts**: Base prompt for build agents (OpenCode default + Sisyphus variants)
|
||||
- **plan-prompt.ts**: Base prompt for plan agents (legacy)
|
||||
- **prometheus-prompt.ts**: System prompt for Prometheus (Planner) agent
|
||||
- **metis.ts**: Metis (Plan Consultant) agent for pre-planning analysis
|
||||
|
||||
Used by `src/index.ts` when creating Builder-Sisyphus and Prometheus (Planner) variants.
|
||||
- **build-prompt.ts**: Unified base for Sisyphus and Builder variants.
|
||||
- **plan-prompt.ts**: Core planning logic shared across planning agents.
|
||||
- **orchestrator-sisyphus.ts**: Uses a 7-section prompt structure and "wisdom notepad" to preserve learnings across turns.
|
||||
|
||||
Reference in New Issue
Block a user