Update submodule changes
All checks were successful
GBCI / build (push) Successful in 12s

This commit is contained in:
Rodrigo Rodriguez 2026-02-13 22:31:47 +00:00
parent 3e418df1a8
commit 6d48dbba1b

131
PROMPT.md
View file

@ -1,15 +1,15 @@
# botbook Development Guide
**Version:** 6.2.0
**Version:** 6.2.0
**Purpose:** Documentation for General Bots (mdBook)
---
## 📚 CRITICAL: Keyword Naming Rules
## CRITICAL: Keyword Naming Rules
**Keywords NEVER use underscores. Always use spaces.**
### Correct Syntax
### Correct Syntax
```basic
SEND MAIL to, subject, body, attachments
GENERATE PDF template, data, output
@ -20,7 +20,7 @@ SET BOT MEMORY key, value
KB STATISTICS
```
### WRONG - Never Use Underscores
### WRONG - Never Use Underscores
```basic
SEND_MAIL ' WRONG!
GENERATE_PDF ' WRONG!
@ -39,7 +39,7 @@ DELETE_HTTP ' WRONG!
---
## 🎨 OFFICIAL ICONS - MANDATORY
## OFFICIAL ICONS - MANDATORY
**NEVER generate icons with LLM. Use official SVG icons from `botui/ui/suite/assets/icons/`**
@ -55,7 +55,7 @@ DELETE_HTTP ' WRONG!
---
## 📁 STRUCTURE
## STRUCTURE
```
botbook/
@ -76,7 +76,7 @@ botbook/
---
## 📖 DOCUMENTATION RULES
## DOCUMENTATION RULES
```
- All documentation MUST match actual source code
@ -88,18 +88,18 @@ botbook/
---
## 🚫 NO ASCII DIAGRAMS — MANDATORY
## NO ASCII DIAGRAMS - MANDATORY
**NEVER use ASCII art diagrams. ALL diagrams must be SVG.**
### Prohibited ASCII Patterns
### Prohibited ASCII Patterns
```
┌─────────┐ ╔═══════╗ +-------+
│ Box │ ║ Box ║ | Box |
└─────────┘ ╚═══════╝ +-------+
+-------+
| Box |
+-------+
```
### What to Use Instead
### What to Use Instead
| Instead of... | Use... |
|---------------|--------|
@ -109,7 +109,7 @@ botbook/
---
## 🎨 SVG DIAGRAM GUIDELINES
## SVG DIAGRAM GUIDELINES
All SVGs must support light/dark modes:
@ -117,7 +117,7 @@ All SVGs must support light/dark modes:
<style>
.title-text { fill: #1E1B4B; }
.main-text { fill: #334155; }
@media (prefers-color-scheme: dark) {
.title-text { fill: #F1F5F9; }
.main-text { fill: #E2E8F0; }
@ -127,7 +127,7 @@ All SVGs must support light/dark modes:
---
## 💬 CONVERSATION EXAMPLES
## CONVERSATION EXAMPLES
Use WhatsApp-style HTML format for bot interactions:
@ -150,19 +150,42 @@ Use WhatsApp-style HTML format for bot interactions:
---
## 📋 SOURCE CODE REFERENCES
## SOURCE CODE REFERENCES
| Topic | Source Location |
|-------|-----------------|
| BASIC Keywords | `botserver/src/basic/keywords/` |
| Database Models | `botserver/src/shared/models.rs` |
| Database Models | `botserver/src/core/shared/models.rs` |
| API Routes | `botserver/src/core/urls.rs` |
| Configuration | `botserver/src/core/config/` |
| Templates | `botserver/templates/` |
---
## 🔧 BUILDING DOCUMENTATION
## BUILDING BOTSERVER
**CRITICAL: ALWAYS USE DEBUG BUILD DURING DEVELOPMENT**
```bash
# CORRECT - Use debug build (FAST)
cargo build
# WRONG - NEVER use --release during development (SLOW)
# cargo build --release # DO NOT USE!
# Run debug server
cargo run
```
**Why Debug Build:**
- Debug builds compile in ~30 seconds
- Release builds take 5-10 minutes with LTO
- Debug builds are sufficient for development and testing
- Only use `--release` for production deployment
---
## BUILDING DOCUMENTATION
```bash
# Install mdBook
@ -177,7 +200,73 @@ mdbook serve --open
---
## 🔑 REMEMBER
## TESTING PROCEDURES
### Tool Testing Workflow
**CRITICAL: NO STOP UNTIL NO MORE ERRORS IN TOOLS**
When testing bot tools, follow this sequential process WITHOUT STOPPING:
```
1. Test Tool #1
├─ Fill form one field at a time (if multi-step form)
├─ Verify NO ERRORS in output
├─ Check Result types are NOT visible (no "core::result::Result<..." strings)
├─ Verify database save (if applicable)
├─ IF ERRORS FOUND: FIX THEM IMMEDIATELY, RE-TEST SAME TOOL
├─ ONLY move to next tool when CURRENT tool has ZERO errors
2. Test Tool #2
└─ (repeat process - DO NOT STOP if errors found)
3. Continue until ALL tools tested with ZERO errors
```
**IMPORTANT:**
- Do NOT stop testing to summarize or ask questions
- Do NOT move to next tool if current tool has errors
- Fix errors immediately, rebuild, re-test same tool
- Only proceed when current tool is completely error-free
### Error Patterns to Watch For
**CRITICAL ERRORS (Must Fix Before Proceeding):**
- `core::result::Result<alloc::string::String, alloc::string::String>` in output
- `invalid input syntax for type timestamp`
- `Some("Desculpe, houve um erro...")`
- Empty balloon messages
- Rust debug info visible to users
### Playwright Testing Tricks
```javascript
// Click tool button
await page.getByRole('button', { name: 'Evento/Iluminação' }).click();
// Wait for response
await page.waitForTimeout(3000);
// Take snapshot
await page.snapshot();
// Fill form field by field
await page.getByRole('textbox').fill('field value');
await page.getByRole('textbox').press('Enter');
```
### Test Documentation
After testing each tool, document:
1. Tool name and ID
2. All required parameters
3. Expected behavior
4. Any issues found and fixes applied
5. Final test result (PASS/FAIL)
---
## REMEMBER
- **Accuracy** - Must match botserver source code
- **Completeness** - No placeholder sections
@ -186,3 +275,5 @@ mdbook serve --open
- **NO ASCII art** - Use SVG diagrams only
- **Version 6.2.0** - Always reference 6.2.0
- **GIT WORKFLOW** - ALWAYS push to ALL repositories (github, pragmatismo)
- **TESTING** - Test tools ONE BY ONE, fix ALL errors before moving to next tool
- **NO STOP** - DO NOT STOP testing until ALL tools have ZERO errors - fix immediately and re-test