PerpetualSoftware/pad: Project Management for the Agent Era

What changed

The PerpetualSoftware/pad repository, a project management tool tailored for the agent era, has seen a recent release, version v0.18.1, on October 18, 2026. This release, occurring just two days before the data snapshot, signifies ongoing development and maintenance.

Key Features and Technologies:

• Core Functionality: Described as "Project Management for the agent era," pad aims to provide tools for task and issue tracking within a local-first, self-hosted architecture.
• Technology Stack: The project is primarily written in Go, with a frontend likely utilizing SvelteKit, as indicated by the presence of sveltekit in its topics. It also leverages SQLite for local data storage and includes Dockerfiles for containerization.
• Agent-Centric Design: Topics like "agent-era," "ai-agents," "claude-code," and "mcp-server" suggest an integration or design philosophy centered around AI agents and their development workflows.
• Developer Focus: The project is positioned as a developer tool, with a CLI interface and support for local-first development, appealing to those who prefer self-hosted and customizable solutions.

Why it matters for builders

For developers and teams working with AI agents, pad offers a specialized toolset that integrates project management with the unique demands of agent-based development. Its local-first and self-hosted nature provides greater control over data and workflows, which is often a requirement in sensitive or complex AI projects. The inclusion of agent-specific topics hints at potential features that could streamline the development, testing, and deployment of AI agents.

Practical impact

Builders can explore pad by cloning the repository and examining its Go codebase and Docker configurations. The recent release suggests that the project is actively maintained, making it a viable option for those seeking an alternative to traditional project management tools. Developers interested in agent development workflows might find pad's specific features, such as its MCP server integration or CLI tooling, particularly useful for managing their agent projects. The project's emphasis on local-first and self-hosted deployment means users can set it up on their own infrastructure for enhanced privacy and customization.

Caveats and source limits

The provided repository metadata is limited in detail regarding the specific functionalities and user experience of pad. While the description and topics offer clues, there is no direct information on installation instructions beyond the presence of Dockerfiles, detailed usage examples, or specific performance benchmarks. The readme_summary mentions "4 AI signals" and "4 developer signals" but does not elaborate on what these entail. Furthermore, documentation availability (hasDocs: false) is explicitly stated as lacking, which may present challenges for new users. The exact nature of its "agent era" features and how they integrate with tools like Claude Code or Codex is not detailed.