Turn a rough request into a structured AI prompt with a clear goal, output format, and constraints. For complex work, add smaller tasks, separate creator and reviewer roles, and explicit checks for accuracy and uncertainty.
Evaluate feedback from reviewers, CI bots, or other AI agents against the actual code and requirements. Apply valid fixes one at a time, verify each change, and explain disagreements using evidence from callers, tests, or existing conventions.
Run repeated rounds of code fixes, performance improvements, or other refinements with defined goals and progress tracking. Evaluate each round, stop when appropriate, and optionally use numeric measurements to keep improvements or roll back regressions.
Restore project files from a previous Git checkpoint after previewing affected files and receiving explicit confirmation. Create a safety tag before restoration, preserve recorded lessons, and keep the current branch and commit history in place.
Finalize a completed Claude Octopus project by checking readiness, running a multi-AI security audit, recording lessons, archiving project state, and creating a Git checkpoint. Produce a delivery summary and mark the project as shipped.
Clarify a feature’s requirements with multiple AI providers, combining technical constraints with user and business needs. Produce a scoped definition covering priorities, edge cases, exclusions, and a requirements checklist before implementation begins.
Review code or documents with available AI providers before delivery. Combine findings into a scored validation report that checks prior requirements and decisions, separates critical issues from warnings, and recommends next steps.
Develop complex features or business documents by gathering approaches from available AI providers. Combine their perspectives with prior project decisions and quality checks to produce a proposed plan, then proceed with implementation after user confirmation.
Create a structured specification in plain language from a feature description and multi-AI research. Define actors, behaviors, constraints, dependencies, and acceptance conditions, while checking completeness and keeping unresolved user decisions visible for later review.
Review system architecture using implementation, callers, and tests as evidence. Compare interface designs, clarify how components should interact, and receive a recommendation with migration steps and test implications before deciding whether to refactor.
Research complex topics across multiple sources using available external AI providers, then combine the findings. Choose the depth, focus, and output format to receive a detailed report, concise summary, comparison table, or recommendations.
Reviews code for common security weaknesses, authentication problems, exposed credentials, and unsafe configuration. It adjusts review depth to the changes and offers an adversarial workflow, where defensive and attacking perspectives guide fixes and subsequent verification.
Plan a new interface using design reference searches, critique, and contrast checks. Produce colors, typography, reusable styling values, component specifications, layouts, and a style guide tailored to the product, audience, and selected technology.
Audit an application or selected features with a structured checklist. Record evidence for broken behavior, incomplete implementations, and inconsistent patterns, then organize findings by severity and propose a prioritized repair plan.
Write or review skills for a consistent process across runs. Check activation conditions, ordered steps, completion criteria, and reference placement, then identify unclear instructions, overlapping triggers, and content that belongs in an existing skill.
Detects whether a task concerns software development or research and strategy by examining the request and project files. Workflow skills can use the result to choose agents, output formats, and review criteria suited to the work.
Is this your account? Link GitHub to bring it onto your creator page (Japanese)