Skill Claude CodeCodex
A tool for starting a new software development task through a complete specification workflow: aligning requirements, exploring the code, designing the solution, implementing it, testing it, and closing the work.
Skill Claude CodeCodex
A coordinator for starting a software development workflow where one main agent delegates all substantive work to spawned child agents. The workflow covers requirements, exploration, design, implementation, testing, and closure.
Skill Claude CodeCodex
A testing workflow for a role called spec-tester. It creates an HTML test plan, runs tests after implementation, and produces an HTML test report; it can also recheck fixes made by a spec-debugger.
Skill Claude CodeCodex
A tool that checks how much context an AI coding assistant receives and finds ways to reduce unnecessary tokens. Tokens are the small text units used to process prompts and tool output.
Skill Claude CodeCodex
A tool for writing technical specifications: documents that describe a software design, its data, APIs, architecture, or planned refactoring. It creates the plan page used by the project’s specification workflow.
Skill Claude CodeCodex
Cross-platform token audit and optimization for AI coding agents (Cursor, Codex, Claude Code, Hermes, Windsurf, Cline, OpenClaw, Gemini CLI, Aider). Scans always-on context, MCP tools, skills; ranks ROI fixes; installs L0-L3 rules; shrinks large tool outputs. Use for save tokens, token audit, context bloat, MCP prune…
Cursor rule
Token efficiency L1 — agent tool-call discipline.
Cursor rule
Token efficiency L0 — context landmines.
Cursor rule
Token efficiency L3 — output discipline.