Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion examples/dynamic-variables/project.promptrek.yaml
Original file line number Diff line number Diff line change
@@ -1,4 +1,4 @@
schema_version: 3.0.0
schema_version: 3.1.0

metadata:
title: Dynamic Variables Example
Expand Down
189 changes: 80 additions & 109 deletions examples/v21-plugins/autonomous-agents.promptrek.yaml
Original file line number Diff line number Diff line change
@@ -1,12 +1,11 @@
# yaml-language-server: $schema=https://promptrek.ai/schema/v2.1.0.json
schema_version: 2.1.0
# yaml-language-server: $schema=https://promptrek.ai/schema/v3.1.0.json
schema_version: 3.1.0
metadata:
title: Autonomous Agents Example
description: Example agent configurations for automated tasks
version: 1.0.0
author: PrompTrek Team
tags: [agents, automation, example]

content: |
# Autonomous Agents Example

Expand All @@ -18,114 +17,86 @@ content: |
- **full**: Agent can act without approval
- **partial**: Agent requires approval for sensitive operations
- **untrusted**: Agent requires approval for all operations
agents:
- name: code-reviewer
prompt: |
You are an expert code reviewer with deep knowledge of software engineering
best practices. Your role is to:

plugins:
agents:
# Code reviewer agent - reviews code automatically
- name: code-reviewer
description: Autonomous code review agent
system_prompt: |
You are an expert code reviewer with deep knowledge of software engineering
best practices. Your role is to:

1. Review code for quality, security, and performance issues
2. Suggest improvements and refactorings
3. Ensure code follows project conventions
4. Identify potential bugs and edge cases
5. Recommend additional tests

Be thorough but constructive in your reviews. Explain the reasoning
behind your suggestions.
tools:
- file_read
- git_diff
- static_analysis
- run_tests
trust_level: partial
requires_approval: true
context:
review_criteria: ["code_quality", "security", "performance", "testing"]
severity_threshold: "medium"
trust_metadata:
trusted: true
trust_level: partial
source: local

# Documentation generator agent
- name: doc-generator
description: Automatically generates and updates documentation
system_prompt: |
You are a documentation specialist. Generate clear, comprehensive
documentation that helps developers understand and use the code.

Focus on:
- Clear explanations
- Practical examples
- Common use cases
- API references
- Troubleshooting guides
tools:
- file_read
- file_write
- analyze_code_structure
trust_level: partial
requires_approval: true
1. Review code for quality, security, and performance issues
2. Suggest improvements and refactorings
3. Ensure code follows project conventions
4. Identify potential bugs and edge cases
5. Recommend additional tests

# Test generator agent
- name: test-generator
description: Generates unit and integration tests
system_prompt: |
You are a test automation expert. Generate comprehensive tests that:
- Cover normal operations
- Test edge cases
- Handle error conditions
- Mock external dependencies
- Follow testing best practices
tools:
- file_read
- file_write
- run_tests
- analyze_coverage
trust_level: partial
requires_approval: true
Be thorough but constructive in your reviews. Explain the reasoning
behind your suggestions.
description: Autonomous code review agent
tools: [file_read, git_diff, static_analysis, run_tests]
trust_level: partial
requires_approval: true
context:
review_criteria: [code_quality, security, performance, testing]
severity_threshold: medium
trust_metadata:
trusted: true
trust_level: partial

# Bug fixer agent - highly autonomous
- name: bug-fixer
description: Automatically fixes simple bugs
system_prompt: |
You are a bug-fixing specialist. Identify and fix bugs while:
- Understanding the root cause
- Applying minimal changes
- Adding tests to prevent regression
- Documenting the fix
tools:
- file_read
- file_write
- git_commit
- run_tests
trust_level: untrusted
requires_approval: true
context:
max_files_per_fix: 3
require_tests: true
auto_commit: false
source: local
- name: doc-generator
prompt: |
You are a documentation specialist. Generate clear, comprehensive
documentation that helps developers understand and use the code.

# Refactoring agent
- name: refactorer
description: Suggests and applies code refactorings
system_prompt: |
You are a refactoring expert. Improve code quality through:
- Extracting functions/classes
- Removing duplication
- Simplifying complex logic
- Improving naming
- Applying design patterns
Focus on:
- Clear explanations
- Practical examples
- Common use cases
- API references
- Troubleshooting guides
description: Automatically generates and updates documentation
tools: [file_read, file_write, analyze_code_structure]
trust_level: partial
requires_approval: true
- name: test-generator
prompt: |
You are a test automation expert. Generate comprehensive tests that:
- Cover normal operations
- Test edge cases
- Handle error conditions
- Mock external dependencies
- Follow testing best practices
description: Generates unit and integration tests
tools: [file_read, file_write, run_tests, analyze_coverage]
trust_level: partial
requires_approval: true
- name: bug-fixer
prompt: |
You are a bug-fixing specialist. Identify and fix bugs while:
- Understanding the root cause
- Applying minimal changes
- Adding tests to prevent regression
- Documenting the fix
description: Automatically fixes simple bugs
tools: [file_read, file_write, git_commit, run_tests]
trust_level: untrusted
requires_approval: true
context:
max_files_per_fix: 3
require_tests: true
auto_commit: false
- name: refactorer
prompt: |
You are a refactoring expert. Improve code quality through:
- Extracting functions/classes
- Removing duplication
- Simplifying complex logic
- Improving naming
- Applying design patterns

Always ensure refactorings maintain existing behavior.
tools:
- file_read
- file_write
- run_tests
- static_analysis
trust_level: untrusted
requires_approval: true
Always ensure refactorings maintain existing behavior.
description: Suggests and applies code refactorings
tools: [file_read, file_write, run_tests, static_analysis]
trust_level: untrusted
requires_approval: true
allow_commands: false
Loading
Loading