tokyo-megacorp
Official@tokyo-megacorp
Offers structured codebase optimization, adversarial quality gate validation, and iterative experiment management for enterprise software development environments.
Agent Skills by tokyo-megacorp
Showing 25 vetted skills indexed across 1 GitHub repositories.
tokyo-megacorpinit
Analyze project files and generate autoimprove.yaml configuration files.
tokyo-megacorpidea-matrix
Evaluate design options with parallel rubric scoring and convergence reports.
tokyo-megacorptrack
Create, monitor, and modify project metric goals in automated code improvement workflows.
tokyo-megacorptest
Execute project quality gates and generate pass/fail reports.
tokyo-megacorpidea-archive
Save idea-matrix convergence reports as markdown files with metadata.
tokyo-megacorpdiff
Inspect code differences between experiment commits.
tokyo-megacorpstatus
Summarize autoimprove session state from state files and git worktree info.
tokyo-megacorpdiagnose
Validate autoimprove.yaml configuration and dry-run gates and benchmarks.
tokyo-megacorpadversarial-review
Coordinate AI agents in adversarial debates to review code quality.
tokyo-megacorpdecisions
Browse, filter, and review archived design decisions in markdown files.
tokyo-megacorpautoimprove
Automate iterative testing, evaluation, and refinement of codebases.
tokyo-megacorprubrics
Evaluate skill assets against structured rubrics for discoverability, robustness, and maintainability.
tokyo-megacorpcleanup
Identify and delete outdated git worktrees and branches.
tokyo-megacorpprompt-testing
Generate automated test scaffolding and validate AI skills with natural language prompts.
tokyo-megacorppolish
Diagnose, review, and improve skill or plugin components for quality.
tokyo-megacorphistory
Filter and review experiment logs by verdict, theme, and date.
tokyo-megacorpexperiment
Create, list, and remove automated code improvement experiments with JSON and Git worktrees.
tokyo-megacorpcalibrate
Compare AI model outputs on adversarial code review tasks to detect reasoning gaps.
tokyo-megacorpreport
Generate comprehensive AutoImprove experiment reports with verdicts, drift, and metric trends.
tokyo-megacorprollback
Revert specific kept experiments using git revert based on experiment logs.
tokyo-megacorprun
Manage autoimprove codebase optimization sessions with setup, execution, and reporting.
tokyo-megacorpchallenge
Benchmark debate agents' bug detection accuracy on curated code challenges with F1 scores.
tokyo-megacorpdocs-regenerate
Apply git diff patches to update affected documentation sections.
tokyo-megacorpproposals
Read proposal files, display details, and record approval decisions with timestamps.
Frequently Asked Questions About tokyo-megacorp
FAQPage SchemaWhat specific tasks can be performed using Tokyo-megacorp?▼
Tokyo-megacorp enables systematic codebase optimization, adversarial code quality reviews, and structured decision-making. Users can execute quality gates, generate rubric-based convergence reports, manage experiment lifecycles via git worktrees, and perform automated benchmarking of code improvements against defined metric goals.
Which personas benefit most from these capabilities?▼
Software engineers, quality assurance leads, and technical architects benefit from these capabilities. The system is designed for teams requiring rigorous, evidence-based code improvement, adversarial validation of logic, and structured tracking of design decisions across complex, iterative development cycles.
What are the prerequisites for running these optimization sessions?▼
Execution requires a git-initialized repository and the presence of an autoimprove.yaml configuration file. The system relies on git worktrees for managing experiment isolation and requires structured rubric definitions to perform valid quality gate assessments and convergence reporting.