Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
This collection provides structured guidance for AI agents to handle complex repository tasks. It enables the precise routing of workflow design, debugging, and upgrade requests to appropriate prompts, ensuring that agentic automation remains consistent.
For developers extending the platform, it streamlines the creation of canvas extensions by automating the scaffolding of plugin manifests and metadata migration.
WHO IT'S FOR
AI agent / automation builders
design and route complex agentic workflows
Technical leads / architects
map and document existing codebase architectures
Platform / DevEx teams
implement agent governance and compliance patterns
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Use the context-map skill for this task: [describe the proposed change]. Analyze the current codebase and produce a context map covering files that may need modification, direct dependencies, related tests, similar existing patterns, and risks. Do not implement the change; stop after presenting the map for review.
Polish an existing prompt without changing its intent
Uses finalize-agent-prompt
Path to an existing prompt file, replacing [path/to/prompt.md]
Use the finalize-agent-prompt skill to review and refine the prompt file at [path/to/prompt.md]. Improve its structure, wording, spelling, grammar, and clarity while preserving its original intent, front matter, encoding, and Markdown structure. Show me the proposed revision for review.
Identify files needed to answer a codebase question
Uses what-context-needed
A specific question about the codebase
Use the what-context-needed skill for this question: [question about the codebase]. Before answering it, list the files you must see, files that would be helpful, any relevant files already available in the conversation, and the remaining uncertainties. Explain why each listed file matters.