How to Get Support for Ponytail: Built-in Commands, GitHub Issues, and Configuration Guides

Start with the /ponytail-help command for instant reference, then consult the README and skills/ponytail-help/SKILL.md for deep configuration details; if issues persist, open a GitHub Issue for direct maintainer support.

Ponytail is a "lazy senior dev" framework designed for AI-agent ecosystems. Whether you are troubleshooting installation errors or customizing default modes, the repository offers structured support channels ranging from built-in CLI skills to community-driven issue tracking.

Built-in Help Skill

The fastest way to resolve questions is through the framework's native help system.

Using the /ponytail-help Command

Ponytail includes a dedicated help skill that serves as a quick-reference card for all modes, skills, and commands. According to the source code in skills/ponytail-help/SKILL.md, you can activate this by typing /ponytail-help or simply asking "what ponytail commands" in your AI console.

# Display the quick-reference card

/ponytail-help

# Switch to ultra mode for the current session

/ponytail ultra

This skill documents every available command and configuration option without requiring you to leave your development environment.

Official Documentation

For comprehensive guidance beyond the inline help, Ponytail maintains detailed documentation in the repository root.

README and Platform Guides

The README.md file serves as the central hub for installation instructions, update procedures, and hook integration. For platform-specific guidance, consult the documents located in the docs/ directory, including platform-native.md and agent-portability.md.

These files cover environment-specific setups and portability considerations across different AI agents, ensuring you can deploy Ponytail regardless of your host platform.

GitHub Issues and Community Support

When documentation does not resolve your problem, the GitHub repository provides direct access to the maintainer and community.

Reporting Bugs and Feature Requests

Open a new issue in the GitHub repository to report bugs or request features. The maintainer actively monitors the issue tracker, as evidenced by historical references such as issue #126 linked in the documentation.

Before filing, check existing issues to see if your problem has already been addressed. The issue tracker also accepts feature requests and usage clarifications.

GitHub Discussions

If your question is more about usage patterns or best practices rather than bugs, use the GitHub Discussions tab. This community-driven forum allows users to share tips, real-world examples, and workarounds for complex integrations.

Updating and Configuring Ponytail

Many support issues resolve simply by updating to the latest version or adjusting configuration parameters.

Automatic Plugin Updates

Keep Ponytail current to receive the latest bug fixes and skills without manual reinstallation. As documented in skills/ponytail-help/SKILL.md (lines 63-68), run the following commands in Claude Code:

/plugin marketplace update ponytail
/reload-plugins

Other hosts may provide equivalent update UIs referenced in their respective documentation sections.

Customizing Default Behavior

Override Ponytail's default full mode or disable auto-activation entirely using environment variables or configuration files. According to skills/ponytail-help/SKILL.md (lines 48-56), you can:

# Set via environment variable (Bash/Zsh)

export PONYTAIL_DEFAULT_MODE=off

Alternatively, edit the JSON configuration file:

// ~/.config/ponytail/config.json
{
  "defaultMode": "off"
}

Valid modes include full, ultra, and off, allowing you to control how aggressively Ponytail participates in your sessions.

Host-Specific Integration Support

Some AI agents require manual configuration to load Ponytail correctly.

Manual Hook Setup for Qoder

For hosts like Qoder that do not auto-load plugins, Ponytail provides hook integration files. Copy hooks/qoder-hooks.json into your Qoder settings as shown in the README (lines 211-218) to enable manual activation.

The command definitions themselves reside in commands/ponytail.toml, which specifies the structure of the /ponytail command and its sub-commands for hosts that parse TOML configuration.

Summary

• Use /ponytail-help for immediate command reference without leaving your editor
• Read README.md and docs/platform-native.md for installation and portability guidance
• File GitHub Issues for bugs, errors, or feature requests directly with the maintainer
• Update regularly using /plugin marketplace update ponytail followed by /reload-plugins
• Configure via PONYTAIL_DEFAULT_MODE environment variable or ~/.config/ponytail/config.json
• Install hooks manually using hooks/qoder-hooks.json for non-auto-loading hosts

Frequently Asked Questions

How do I access the Ponytail help documentation instantly?

Type /ponytail-help in your AI console or ask "what ponytail commands" to display the quick-reference card defined in skills/ponytail-help/SKILL.md. This provides immediate access to all available modes and configuration options without browsing external websites.

Where do I report bugs or request new features for Ponytail?

Open a GitHub Issue on the DietrichGebert/ponytail repository. The maintainer actively monitors the issue tracker, and community members often contribute solutions to existing threads. For usage questions rather than bugs, use the GitHub Discussions tab.

How do I disable Ponytail's automatic activation?

Set the PONYTAIL_DEFAULT_MODE environment variable to off or create/edit ~/.config/ponytail/config.json to include "defaultMode": "off". This prevents Ponytail from auto-activating in new sessions while keeping the plugin installed for manual use.

What file contains the command definitions for Ponytail?

The primary command structure is defined in commands/ponytail.toml, which specifies the /ponytail command and its sub-commands. For help-specific commands, see skills/ponytail-help/SKILL.md, which implements the /ponytail-help logic and documents configuration parameters.

{
  "mcpServers": {
    "instagit": {
      "command": "npx",
      "args": ["-y", "instagit@latest"]
    }
  }
}