Install
$ agentstack add mcp-jorgedlt-modeforge ✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
About
ModeForge
[](https://github.com/jorgedlt/modeforge/actions/workflows/validate.yml) [](https://opensource.org/licenses/MIT)
A public collection of pre-built OpenCode mode templates (starter packs) that enable users to quickly deploy time-based, triggered, and workflow-oriented automation using OpenCode and MCP tools.
Overview
OpenCode modes are specialized configurations that define how the AI agent behaves for specific tasks, including custom prompts, temperature settings, tool access, and automation triggers. ModeForge provides ready-to-use JSON definitions to streamline setup and deployment of automated workflows.
Features
- 10 ready-to-run mode templates for common automation scenarios
- Bash and MCP integration examples for extended functionality
- Scheduling and webhook examples for time-based and event-driven execution
- Extensible JSON structure for easy customization and expansion
Quick Start
git clone https://github.com/jorgedlt/modeforge.git
cd modeforge
./scripts/run_mode.sh modes/high-precision.json
For detailed usage instructions, see [HOWTOUSE.md](HOWTOUSE.md).
For detailed usage instructions, see [HOWTOUSE.md](HOWTOUSE.md).
Mode List Summary
- Scheduled Report Mode: Runs on cron, generates summaries and reports from logs or data.
- Webhook Trigger Mode: Listens for API calls, runs specific commands or workflows on trigger.
- High Precision Mode: Low temperature setting for detailed, accurate reasoning tasks.
- Low Cost Mode: Uses cheaper models optimized for bulk processing and repetitive tasks.
- Review Mode: Performs code or document reviews with focused analysis tools.
- Branching Skeleton: AI decision tree runner for conditional workflows and branching logic.
- Screen Capture Stub: Integrates with external vision MCP tools for image-based automation.
- Build/CI Agent Mode: Automates build, test, and deployment cycles in CI/CD pipelines.
- Data Analysis Mode: Handles CSV/SQL data insights, queries, and visualization.
- Maintenance Mode: Runs cleanup tasks, log reviews, and system maintenance operations.
Extending Modes
Modes are defined in JSON format with the following structure. Modify fields to customize behavior:
{
"name": "scheduled-report",
"prompt": "Summarize logs and produce a weekly report.",
"temperature": 0.3,
"tools": ["shell", "git", "http"],
"run": {
"schedule": "0 9 * * MON",
"trigger": "cron"
}
}
name: Unique identifier for the modeprompt: System prompt defining the agent's behaviortemperature: Controls randomness (0.0-1.0, lower for precision)tools: Array of enabled tools (e.g., "shell", "git", "http", MCP tools)run: Automation configuration with schedule (cron format) and trigger type
Contributing
We welcome contributions! To propose new mode types or improve existing ones:
- Fork the repository
- Create a feature branch (
git checkout -b new-mode-template) - Add your mode JSON file to
modes/ - Update this README with the new mode description
- Submit a pull request with a clear description of the changes
Please ensure all JSON files are syntactically valid and include comprehensive documentation.
License
This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.
Source & license
This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: jorgedlt
- Source: jorgedlt/modeforge
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet — be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.