Files
tendril/docs/PROMPTS/Prompt-Creation-Guide.md
Gitea Actions 6a4683022f docs(prompts): implement Phase 4 LLM Usage Guides Setup
This commit implements the complete LLM Usage Guides system for the Tendril
project, establishing reusable prompt documents for AI-assisted workflows.

## What Was Implemented

### 1. Core Guide Documents

**docs/PROMPTS/LLM-Usage-Guide.md** - Comprehensive LLM execution instructions:
- Recognizing prompt documents (naming convention, structure, location)
- Step-by-step execution process (5 phases: recognize, gather info, execute, handle errors, validate)
- Information gathering requirements and best practices
- Error handling procedures and common error patterns
- Validation steps (pre-execution, during execution, post-execution)
- Providing feedback (progress updates, error reporting, completion summaries)
- Best practices for execution, communication, and quality
- Special instructions for Tendril-specific context
- Troubleshooting guide for common prompt execution issues
- Customized for Tendril project structure and Gitea platform

**docs/PROMPTS/Prompt-Creation-Guide.md** - Guide for creating effective prompts:
- Naming convention (NN-Descriptive-Name-Prompt.md format)
- Complete prompt structure template with all required sections
- Best practices for writing effective instructions
- Information gathering guidelines
- Error handling documentation standards
- Testing procedures for new prompts
- Common patterns (status check, search, sync patterns)
- Decision criteria for when to create new prompts
- Update procedures for existing prompts
- Comprehensive checklist for new prompts
- Examples customized for Tendril project
- References to Gitea-specific workflows and features

### 2. Initial Prompt Documents

**docs/PROMPTS/00-Project-Status-Check-Prompt.md** - Project status overview:
- Purpose: Check current status of all project phases
- Reads all phase blueprints in tendril/phases/
- Extracts phase status, deliverables, milestones
- Organizes by status (Complete, In Progress, Planned)
- Provides comprehensive status summary with counts
- Highlights issues or blockers
- Error handling for missing files, invalid formats
- Validation checklist for completion
- Customized for Tendril phase structure

**docs/PROMPTS/01-KB-Content-Search-Prompt.md** - KB system search:
- Purpose: Search KB system for relevant content
- Supports multiple search types: topics, tags, phase relevance, keywords
- Uses kb/_index.md for efficient searching
- Reads matching KB files and extracts metadata
- Presents organized search results with summaries
- Handles missing index, invalid criteria, empty results
- Provides helpful suggestions for refining searches
- Customized for Tendril KB structure and categories

**docs/PROMPTS/02-Phase-Documentation-Sync-Prompt.md** - Phase documentation synchronization:
- Purpose: Ensure phase documentation consistency
- Reads all four phase documents (blueprint, tasks, decisions, changelog)
- Compares and identifies discrepancies
- Syncs tasks.md with blueprint deliverables and Definition of Done
- Updates changelog.md with sync entry
- Reviews decisions.md for missing ADRs
- Verifies consistency across all documents
- Follows Cursor rules for phase documentation synchronization
- Error handling for missing files, format issues, unresolvable discrepancies
- Customized for Tendril phase structure and documentation system

## Why This Implementation

### Standardized AI Workflows
   LLM Usage Guides enable consistent, repeatable AI-assisted workflows. Instead
   of explaining the same tasks repeatedly, prompts provide standardized
   instructions that work reliably every time.

### Reusability and Efficiency
   Prompt documents can be dragged into Cursor chat sessions and executed
   immediately. This saves time and ensures tasks are performed consistently
   across different sessions and users.

### Quality and Reliability
   Well-structured prompts include validation checklists, error handling, and
   best practices. This reduces errors and ensures high-quality results.

### Knowledge Preservation
   Prompts document how to perform complex tasks, preserving institutional
   knowledge and making it accessible to anyone who needs to perform the task.

### Integration with Project Systems
   Prompts are designed to work with Tendril's phase documentation system, KB
   system, and Gitea workflows, providing seamless integration with existing
   project infrastructure.

## Technical Details

### Prompt Document Structure

All prompts follow a consistent structure:
1. **Title and Purpose** - Clear description of what the prompt does
2. **How to Use** - Instructions for using the prompt
3. **CRITICAL: Information Gathering** - Required information section
4. **Step-by-Step Process** - Phased execution instructions
5. **Error Handling** - Common issues and solutions
6. **Validation Checklist** - Completion verification
7. **Important Notes** - Critical instructions for AI assistant

### Naming Convention

Prompts follow strict naming:
- 00-09: Core system prompts
- 10-19: Phase management prompts
- 20-29: KB system prompts
- 30-39: Workflow and automation prompts
- 40-49: Documentation prompts
- 50+: Project-specific prompts

### Customization for Tendril

All prompts and guides are customized for:
- **Project Name**: Tendril (not "pairs" or placeholders)
- **Platform**: Gitea (not GitHub)
- **Repository**: https://git.parkingmeter.info/Mycelium/tendril
- **Directory Structure**: tendril/phases/, kb/ with category directories
- **Workflows**: Gitea Actions (with GitHub Actions compatibility notes)

### Integration Points

Prompts integrate with:
- **Phase Documentation**: tendril/phases/phase-XX-name/ files
- **KB System**: kb/ directory with index and category structure
- **Workflows**: .github/workflows/ (Gitea Actions)
- **Documentation**: docs/ directory structure

## Files Added

- docs/PROMPTS/LLM-Usage-Guide.md - LLM execution instructions (~400 lines)
- docs/PROMPTS/Prompt-Creation-Guide.md - Prompt creation guide (~600 lines)
- docs/PROMPTS/00-Project-Status-Check-Prompt.md - Status check prompt (~180 lines)
- docs/PROMPTS/01-KB-Content-Search-Prompt.md - KB search prompt (~240 lines)
- docs/PROMPTS/02-Phase-Documentation-Sync-Prompt.md - Documentation sync prompt (~250 lines)

**Total**: 5 files, ~1,670 lines of prompt documentation

## Validation

All Phase 4 requirements met:
-  LLM Usage Guide created and comprehensive
-  Prompt Creation Guide created with complete template
-  Three initial prompt documents created
-  All prompts follow naming convention
-  All prompts include required sections
-  All files reference "Tendril" (42 references found)
-  All files use "Gitea" terminology (10 references, no GitHub)
-  Examples relevant to Tendril project structure
-  Documentation is clear and actionable

## Next Steps

Phase 4 complete. LLM Usage Guides system is ready for use. Prompts can be
dragged into Cursor chat sessions to perform standardized tasks. Ready for
Phase 5: Validation and Testing.

## Related

- Phase 4 Plan: kb/_guides/PROJECT-SETUP-GUIDE/PHASE-4-PLAN.md
- Replication Guide: kb/_guides/PROJECT-SETUP-GUIDE/COMPLETE-SYSTEM-REPLICATION-GUIDE.md
- Gitea LLM Guidelines: docs/GITEA/LLM-Gitea-Guidelines.md
2025-11-11 12:25:52 -07:00

14 KiB

Prompt Creation Guide

Purpose: Guide for creating effective, reusable prompt documents for AI-assisted workflows in the Tendril project.

Last Updated: 2025-11-11


Overview

This guide provides step-by-step instructions for creating prompt documents that enable consistent, reliable AI-assisted workflows. Prompt documents are reusable instructions that can be dragged into Cursor chat sessions to accomplish specific tasks.

Key Principles:

  • Clarity: Instructions must be unambiguous
  • Completeness: All required information must be specified
  • Reusability: Prompts should work for similar tasks
  • Validation: Include checklists to verify completion

Naming Convention

Format

Prompt documents must follow this naming pattern:

NN-Descriptive-Name-Prompt.md

Where:

  • NN is a zero-padded two-digit number (00, 01, 02, etc.)
  • Descriptive-Name uses kebab-case (lowercase with hyphens)
  • Always ends with -Prompt.md

Examples

  • 00-Project-Status-Check-Prompt.md
  • 01-KB-Content-Search-Prompt.md
  • 02-Phase-Documentation-Sync-Prompt.md
  • 10-Create-New-Phase-Prompt.md
  • 20-Update-KB-Entry-Prompt.md

Numbering Strategy

  • 00-09: Core system prompts (status checks, searches, syncs)
  • 10-19: Phase management prompts
  • 20-29: KB system prompts
  • 30-39: Workflow and automation prompts
  • 40-49: Documentation prompts
  • 50+: Project-specific prompts

Prompt Structure Template

Every prompt document must include these sections in order:

1. Title and Purpose

# [Task Name] Prompt

**Purpose:** [One sentence describing what this prompt accomplishes]

**How to Use:** Drag this entire document into a Cursor chat session, then [provide context/information]. The AI assistant will follow these instructions to [accomplish the task].

2. CRITICAL: Information Gathering

## CRITICAL: Information Gathering

### Required Information (You should provide this)

Before starting, **ASK THE USER** for the following information if not already provided:

1. **[Information Item 1]** (REQUIRED)
   - Format/constraints
   - Examples
   - Why it's needed

2. **[Information Item 2]** (REQUIRED)
   - Format/constraints
   - Examples
   - Why it's needed

### If Information is Missing

**ALWAYS ASK** for missing required information before proceeding. Do not make assumptions about:
- [List of things not to assume]

3. Step-by-Step Process

## Step-by-Step Process

### Phase 1: [Phase Name]

[Description of what this phase accomplishes]

1. **Step description:**
   ```bash
   # Commands or code
  1. Validation:
    • How to verify this step succeeded
    • What to check

Phase 2: [Phase Name]

[Description of what this phase accomplishes]

  1. Step description: [Instructions]

  2. Validation:

    • How to verify this step succeeded
    • What to check

### 4. Error Handling

```markdown
## Error Handling

### Common Issues

#### [Issue Name]
- **Error message**: What you'll see
- **Fix**: How to resolve it
- **Prevention**: How to avoid it

#### [Another Issue]
- **Error message**: What you'll see
- **Fix**: How to resolve it
- **Prevention**: How to avoid it

5. Validation Checklist

## Validation Checklist

Before completing, verify:

- [ ] [Checklist item 1]
- [ ] [Checklist item 2]
- [ ] [Checklist item 3]
- [ ] ...

6. Important Notes for AI Assistant

## Important Notes for AI Assistant

1. **[Critical instruction 1]**
2. **[Critical instruction 2]**
3. **[Critical instruction 3]**
4. ...

Best Practices

Writing Effective Instructions

  1. Be Specific

    • Use exact commands and paths
    • Specify file names and locations
    • Include expected outputs
  2. Be Complete

    • Include all necessary steps
    • Don't assume prior knowledge
    • Provide context for each step
  3. Be Clear

    • Use simple, direct language
    • Avoid ambiguity
    • Use examples where helpful
  4. Be Structured

    • Organize steps logically
    • Group related steps into phases
    • Number steps sequentially
  5. Be Validated

    • Include validation for each step
    • Provide completion checklist
    • Specify success criteria

Information Gathering

  1. List All Requirements

    • Don't assume any information
    • Be explicit about what's needed
    • Provide format examples
  2. Explain Why

    • Help users understand why information is needed
    • Provide context for requirements
    • Show how information will be used
  3. Provide Examples

    • Give sample values
    • Show format requirements
    • Include valid/invalid examples

Error Handling

  1. Anticipate Common Errors

    • Think about what could go wrong
    • Document likely errors
    • Provide clear fixes
  2. Include Prevention

    • Explain how to avoid errors
    • Provide best practices
    • Note common pitfalls
  3. Provide Clear Fixes

    • Step-by-step error resolution
    • Explain what each fix does
    • Verify fix effectiveness

Writing Effective Instructions

Command Instructions

When providing commands:

1. **Run the command:**
   ```bash
   git status
  1. Verify output:
    • Should show list of modified files
    • Should indicate current branch
    • Should show no uncommitted changes (if clean)

### File Operations

When working with files:

```markdown
1. **Read the file:**
   - Path: `tendril/phases/phase-01-name/blueprint.md`
   - Extract: Project name, status, deliverables

2. **Update the file:**
   - Modify: Status field
   - Add: New deliverable entry
   - Verify: Format is correct

Multi-Step Processes

When breaking down complex tasks:

### Phase 1: Preparation

1. **Gather information:**
   - [What to gather]
   - [Where to find it]
   - [How to verify it]

2. **Validate prerequisites:**
   - [Check 1]
   - [Check 2]
   - [Check 3]

### Phase 2: Execution

1. **Perform action:**
   - [Step 1]
   - [Step 2]
   - [Step 3]

2. **Verify results:**
   - [Verification 1]
   - [Verification 2]

Testing Your Prompt

Test Process

  1. Create the prompt document

    • Follow the structure template
    • Include all required sections
    • Customize for your task
  2. Test with a simple case

    • Use minimal, clear inputs
    • Verify all steps work
    • Check validation checklist
  3. Test with edge cases

    • Missing information
    • Invalid inputs
    • Error conditions
  4. Refine based on results

    • Add missing steps
    • Clarify ambiguous instructions
    • Improve error handling

Testing Checklist

  • Prompt executes successfully with valid inputs
  • Prompt handles missing information correctly
  • Prompt provides clear error messages
  • Validation checklist is complete
  • All steps are clear and unambiguous
  • Examples are helpful and accurate

Common Patterns

Status Check Pattern

# [System] Status Check Prompt

**Purpose:** Check current status of [system/component]

## Step-by-Step Process

### Phase 1: Gather Current State
1. Read [status file]
2. Extract [status indicators]
3. List [current items]

### Phase 2: Analyze Status
1. Compare with [baseline/expected]
2. Identify [issues/discrepancies]
3. Summarize findings

### Phase 3: Report
1. Present status summary
2. Highlight issues
3. Suggest next steps

Search Pattern

# Search [Content Type] Prompt

**Purpose:** Search [system] for [content type] matching [criteria]

## CRITICAL: Information Gathering

1. **Search criteria** (REQUIRED)
   - Format: [format]
   - Examples: [examples]

## Step-by-Step Process

### Phase 1: Prepare Search
1. Read [index/reference]
2. Identify [search targets]
3. Prepare [search parameters]

### Phase 2: Execute Search
1. Search by [method 1]
2. Search by [method 2]
3. Combine results

### Phase 3: Present Results
1. List matches
2. Summarize findings
3. Provide references

Sync Pattern

# Synchronize [Documents] Prompt

**Purpose:** Ensure [documents] are synchronized and consistent

## Step-by-Step Process

### Phase 1: Read All Documents
1. Read [document 1]
2. Read [document 2]
3. Read [document 3]

### Phase 2: Compare and Identify Discrepancies
1. Compare [field 1] across documents
2. Compare [field 2] across documents
3. List all discrepancies

### Phase 3: Update Documents
1. Update [document 1] with [changes]
2. Update [document 2] with [changes]
3. Verify consistency

When to Create a New Prompt

Create a Prompt When:

  1. Task is Repetitive

    • You find yourself doing the same task multiple times
    • Task has clear, repeatable steps
    • Task would benefit from standardization
  2. Task is Complex

    • Task involves multiple phases
    • Task requires specific information gathering
    • Task has validation requirements
  3. Task Needs Consistency

    • Different people perform the task
    • Task must follow specific rules
    • Task has quality requirements
  4. Task Has Common Errors

    • Task frequently encounters errors
    • Errors have known solutions
    • Error handling would be helpful

Don't Create a Prompt When:

  1. Task is One-Time

    • Task will only be done once
    • Task is highly specific
    • Task doesn't need standardization
  2. Task is Too Simple

    • Task is a single command
    • Task requires no validation
    • Task has no error handling needs
  3. Task is Too Variable

    • Task changes significantly each time
    • Task has no clear pattern
    • Task can't be standardized

Updating Existing Prompts

When to Update

  1. Process Changes

    • Workflow has changed
    • New steps are required
    • Steps are no longer needed
  2. Error Discovery

    • New common errors found
    • Existing fixes don't work
    • Prevention methods needed
  3. Clarity Improvements

    • Instructions are ambiguous
    • Examples are unclear
    • Validation is incomplete

Update Process

  1. Test Current Prompt

    • Verify it still works
    • Identify issues
    • Note improvements needed
  2. Make Updates

    • Update affected sections
    • Maintain structure
    • Preserve working parts
  3. Test Updated Prompt

    • Verify updates work
    • Check all sections
    • Validate completeness
  4. Document Changes

    • Note what changed
    • Explain why
    • Update version/date if applicable

Checklist for New Prompts

Before considering a prompt complete, verify:

Structure

  • Title follows naming convention
  • Purpose is clear and concise
  • "How to Use" section is present
  • Information gathering section is complete
  • Step-by-step process is detailed
  • Error handling section exists
  • Validation checklist is present
  • Important notes section included

Content

  • All required information is listed
  • Information formats are specified
  • Examples are provided
  • Steps are clear and unambiguous
  • Validation criteria are specified
  • Common errors are documented
  • Error fixes are provided

Quality

  • Instructions are testable
  • Prompt has been tested
  • Examples are accurate
  • Validation checklist is complete
  • Error handling is comprehensive
  • Customized for Tendril project
  • Uses Gitea terminology (not GitHub)

Tendril-Specific

  • References "Tendril" (not "pairs")
  • Uses Gitea terminology
  • References correct repository URL
  • Uses Tendril directory structure
  • References Tendril-specific paths

Examples

Example 1: Simple Status Check

# Project Status Check Prompt

**Purpose:** Check current status of all project phases

**How to Use:** Drag this document into Cursor chat. The AI will read all phase blueprints and provide a status summary.

## CRITICAL: Information Gathering

No additional information required. The AI will read phase blueprints automatically.

## Step-by-Step Process

### Phase 1: Read Phase Blueprints
1. Find all `blueprint.md` files in `tendril/phases/`
2. Read each blueprint
3. Extract: phase name, status, key deliverables

### Phase 2: Summarize Status
1. List all phases with status
2. Count: Complete, In Progress, Planned
3. Identify any issues or blockers

## Validation Checklist
- [ ] All phase blueprints read
- [ ] Status summary provided
- [ ] Phase counts are accurate

Example 2: Complex Sync Task

# Phase Documentation Sync Prompt

**Purpose:** Ensure phase documentation (blueprint, tasks, decisions, changelog) is synchronized

## CRITICAL: Information Gathering

1. **Phase name or path** (REQUIRED)
   - Format: `phase-XX-name` or full path
   - Example: `phase-01-foundation` or `tendril/phases/phase-01-foundation/`
   - Why: Need to know which phase to sync

## Step-by-Step Process

### Phase 1: Read All Documents
1. Read `blueprint.md`
2. Read `tasks.md`
3. Read `decisions.md`
4. Read `changelog.md`

### Phase 2: Compare and Identify Discrepancies
1. Compare project name across all documents
2. Compare status indicators
3. Compare deliverables/tasks
4. List all discrepancies

### Phase 3: Update Documents
1. Update `tasks.md` to match blueprint
2. Update `changelog.md` with sync entry
3. Review `decisions.md` for missing ADRs
4. Verify all documents are consistent

## Error Handling

### Phase Directory Not Found
- **Error**: Directory doesn't exist
- **Fix**: Verify phase name/path is correct
- **Prevention**: Use exact phase name from phases directory

## Validation Checklist
- [ ] All four documents read
- [ ] Discrepancies identified
- [ ] Documents updated
- [ ] Consistency verified

Summary

Creating effective prompt documents requires:

  1. Follow the structure template - Include all required sections
  2. Be specific and complete - Don't leave room for interpretation
  3. Include validation - Verify completion at each step
  4. Handle errors - Anticipate and document common issues
  5. Test thoroughly - Verify prompt works before using
  6. Customize for Tendril - Use project-specific context

Remember: Good prompts are reusable, reliable, and reduce errors. Invest time in creating quality prompts to save time in the long run.


Location: docs/PROMPTS/Prompt-Creation-Guide.md
Related:

  • docs/PROMPTS/LLM-Usage-Guide.md
  • docs/GITEA/LLM-Gitea-Guidelines.md
  • kb/_guides/PROJECT-SETUP-GUIDE/PHASE-4-PLAN.md