Convert state.json agent execution traces into interactive HTML decision-tree reports.
npx skills add https://github.com/JotJunior/cstk --skill decision-tree-jotjunior
Please help me install this Agent Skill. Skill: decision-tree Source: https://github.com/JotJunior/cstk/tree/main/global/skills/decision-tree Command: npx skills add https://github.com/JotJunior/cstk --skill decision-tree-jotjunior
This skill generates a navigable, self-contained visualization of an execution's decision history from a state.json, enabling quick understanding of IA outcomes.
Run the render script with the path to your state.json and optional output file.
High-intent search queries and answers about installing and using this skill.
You can visualize agent decision history from a state.json file by running the render script, which uses jq to extract data and generates a standalone interactive HTML decision-tree report for offline review.
Debugging agent-00c execution flows is best achieved by converting the state.json decision trace into a navigable HTML tree visualization, enabling quick understanding and auditing of IA outcomes.
Yes, you can generate a completely self-contained HTML report from a state.json decision trace. This standalone file can be opened offline to review the interactive decision-tree.
Yes, jq is required. The skill depends on jq to read the state.json file and extract the execution decision data required to render the HTML visualization.
No, visualizing a state.json decision trace never modifies the source state. The process strictly reads the file to render the HTML report, preserving the original execution data.
EntityProcessAnalyze AgentV evaluation traces with CLI commands and jq.
AJBcodingCreate custom interactive D3.js visualizations with data binding and responsive sizing.
JotJuniorManage state, locking, and secrets filtering for the agente-00c runtime.
Aken-TeamsGenerate a self-contained HTML dashboard with inline charts and CDN libraries.
