Ikatsuba

This skill analyzes a codebase issue and proposes concrete, actionable solutions to fix bugs, improve performance, or address unexpected behavior.

This skill helps you define a comprehensive project vision, consolidating goals, stakeholders, constraints, and shared architectural decisions for large

This skill analyzes project specs to generate a live status dashboard with progress, dependencies, and next-action recommendations.

This skill performs a comprehensive security audit of AI agents, auditing lethal trifecta, sandboxing, access control, and guardrails to prevent breaches.

This skill helps you design and optimize prompts, select models, and craft few-shot examples for reliable, cost-efficient production AI agents.

This skill designs agent memory strategy across sessions, guiding memory layers, processors, and context management to improve consistency.

This skill guides you through designing agent tools, schemas, and MCP strategy to build scalable, reusable capabilities for AI agents.

This skill helps you design graph-based agent workflows with branching, chaining, suspend/resume, streaming, and observability for reliable AI tasks.

This skill guides you through designing multi-agent systems with supervisor patterns, control flow, and workflow composition for scalable collaboration.

This skill executes tests from a specification's test plan, guiding you through all, next, or specific tests with results and updates.

This skill analyzes code changes before commit, providing structured feedback to improve quality, security, and maintainability.

This skill creates a detailed implementation plan by analyzing requirements and design to produce a tracked task breakdown.

This skill executes all tests from a spec plan, stepping through each pending case and recording results for full manual testing.

This skill executes a specified test from a plan, enabling targeted re-testing and verification of fixes.

This skill executes the next pending task from a specification's tasks document, guiding incremental implementation with clear task-by-task progress.

This skill executes all pending tasks from a spec, running major tasks sequentially and enabling parallel subtasks when safe.