- Introduced agent instruction template for generating documentation tasks. - Created CSV schema for tasks including definitions and example entries. - Defined documentation dimensions across five waves with specific focus areas. - Established quality standards for documentation, including completeness, accuracy, clarity, and context integration.
7.1 KiB
Agent Instruction Template (Optimized)
Enhanced instruction template with mandatory discovery check and wave summary integration.
Purpose
| Phase | Usage |
|---|---|
| Wave Execution | Injected as instruction parameter to spawn_agents_on_csv |
Optimized Instruction Template
## DOCUMENTATION TASK — Wave {wave_number}
### ⚠️ MANDATORY FIRST STEPS (DO NOT SKIP)
**CRITICAL**: Complete these steps BEFORE reading any source files!
1. **🔍 CHECK DISCOVERIES FIRST** (避免重复工作):
```bash
# Search for existing discoveries about your topic
cat {session_folder}/discoveries.ndjson | grep -i "{doc_type}"
cat {session_folder}/discoveries.ndjson | grep -i "component\|algorithm\|pattern"
What to look for:
- Already discovered components in your scope
- Existing pattern definitions
- Pre-documented algorithms
-
📊 Read Wave Summary (高密度上下文):
- File: {session_folder}/wave-summaries/wave-{prev_wave}-summary.md
- This contains synthesized findings from previous wave
- USE THIS - it's pre-digested context!
-
📋 Read prev_context (provided below)
Your Task
Task ID: {id} Title: {title} Document Type: {doc_type} Target Scope: {target_scope} Required Sections: {doc_sections} LaTeX Support: {formula_support} Priority: {priority}
Task Description
{description}
Previous Context (USE THIS!)
{prev_context}
Execution Protocol
Step 1: Discovery Check (MANDATORY - 2 min max)
Before reading ANY source files, check discoveries.ndjson:
| Discovery Type | What It Tells You | Skip If Found |
|---|---|---|
component_found |
Existing component | Reuse description |
pattern_found |
Design patterns | Reference existing |
algorithm_found |
Core algorithms | Don't re-analyze |
api_found |
API signatures | Copy from discovery |
Goal: Avoid duplicating work already done by previous agents.
Step 2: Scope Analysis
Read files matching {target_scope}:
- Identify key structures, functions, classes
- Extract relevant code patterns
- Note file:line references for examples
Step 3: Context Integration
Synthesize from multiple sources:
- Wave Summary: High-density insights from previous wave
- prev_context: Specific findings from context_from tasks
- Discoveries: Cross-cutting findings from all agents
Step 4: Document Generation
Determine Output Path by doc_type:
| doc_type | Output Directory |
|---|---|
overview |
docs/01-overview/ |
architecture |
docs/02-architecture/ |
implementation |
docs/03-implementation/ |
theory |
docs/03-implementation/ |
feature |
docs/04-features/ |
api |
docs/04-features/ |
usage |
docs/04-features/ |
synthesis |
docs/05-synthesis/ |
Document Structure (ALL sections required):
# {Title}
## Overview
[Brief introduction - 2-3 sentences]
## {Required Section 1}
[Content with code examples and file:line references]
## {Required Section 2}
[Content with Mermaid diagrams if applicable]
... (repeat for ALL sections in doc_sections)
## Code Examples
```{language}
// src/module/file.ts:42-56
function example() {
// ...
}
Diagrams (if applicable)
graph TD
A[Component] --> B[Dependency]
Cross-References
- Related: Document
- Depends on: Prerequisite
- See also: Reference
Summary
[3-5 key takeaways in bullet points]
### Step 5: Share Discoveries (MANDATORY)
After completing analysis, share findings:
```bash
echo '{"ts":"<ISO8601>","worker":"{id}","type":"<TYPE>","data":{...}}' >> {session_folder}/discoveries.ndjson
Discovery Types (use appropriate type):
| Type | When to Use | Data Fields |
|---|---|---|
component_found |
Found a significant module/class | {name, type, file, purpose} |
pattern_found |
Identified design pattern | {pattern_name, location, description} |
algorithm_found |
Core algorithm identified | {name, file, complexity, purpose} |
formula_found |
Mathematical formula (theory docs) | {name, latex, file, context} |
feature_found |
User-facing feature | {name, entry_point, description} |
api_found |
API endpoint or function | {endpoint, file, parameters, returns} |
config_found |
Configuration option | {name, file, type, default_value} |
Share at least 2 discoveries per task.
Step 6: Report Results
{
"id": "{id}",
"status": "completed",
"findings": "Structured summary (max 500 chars). Include: 1) Main components found, 2) Key patterns, 3) Critical insights. Format for easy parsing by next wave.",
"doc_path": "docs/XX-category/filename.md",
"key_discoveries": "[{\"name\":\"ComponentA\",\"type\":\"class\",\"description\":\"Handles X\",\"file\":\"src/a.ts:10\"}]",
"error": ""
}
Quality Checklist
Before reporting complete, verify:
- All sections present: Every section in
doc_sectionsis included - Code references: Include
file:linefor code examples - Discovery sharing: At least 2 discoveries added to board
- Context usage: Referenced findings from prev_context or Wave Summary
- Cross-references: Links to related documentation
- Summary: Clear key takeaways
LaTeX Formatting (when formula_support=true)
Use $$...$$ for display math:
The weak form is defined as:
$$
a(u,v) = \int_\Omega \nabla u \cdot \nabla v \, d\Omega
$$
Where:
- $u$ is the trial function
- $v$ is the test function
Tips for Effective Documentation
- Be Precise: Use specific file:line references
- Be Concise: Summarize key points, don't copy-paste entire files
- Be Connected: Reference related docs and discoveries
- Be Structured: Follow the required section order
- Be Helpful: Include practical examples and use cases
---
## Placeholder Reference
| Placeholder | Resolved By | Description |
|-------------|-------------|-------------|
| `{wave_number}` | Wave Engine | Current wave number |
| `{session_folder}` | Wave Engine | Session directory path |
| `{prev_wave}` | Wave Engine | Previous wave number (wave_number - 1) |
| `{id}` | CSV row | Task ID |
| `{title}` | CSV row | Document title |
| `{doc_type}` | CSV row | Document type |
| `{target_scope}` | CSV row | File glob pattern |
| `{doc_sections}` | CSV row | Required sections |
| `{formula_support}` | CSV row | LaTeX support flag |
| `{priority}` | CSV row | Task priority |
| `{description}` | CSV row | Task description |
| `{prev_context}` | Wave Engine | Aggregated context |
---
## Optimization Notes
This instruction template includes:
1. **Mandatory Discovery Check**: Forces agents to check discoveries.ndjson first
2. **Wave Summary Integration**: References previous wave's synthesized findings
3. **Structured Reporting**: Findings formatted for easy parsing by next wave
4. **Quality Checklist**: Explicit verification before completion
5. **Discovery Requirements**: Minimum 2 discoveries per task