Turn review into an organizational judgment asset. Review Judgment as Code for AI-assisted development, delivered as a capability pack for Claude Code: your team's review criteria as versioned, repo-owned skills (a Skill Registry), driven by a review agent that runs perspective-based reviewer roles (code, security…
★not rated 7▲
+3
changed todayA
tokens not measured
originalMIT
Cursor rule "cursorrules" from s977043/river-review, covering river reviewer — cursor agent rules, npm i -g @s977043/river-review, verify: river --version, https://river-reviewer.s977043.dev/docs/quickstart and when to run river reviewer.
A Japanese-language workflow for converting requirements and edge cases into acceptance criteria. Acceptance criteria are checkable conditions that define when a feature is finished.
A design-planning aid that turns an agreed specification into a proposed software structure, including modules, data movement, state, failures, and testing.
A conversational design process that turns a rough feature idea or unclear requirement into an approved design document. It asks focused questions and considers several implementation approaches before code is written.
A guide for dividing development work among multiple coding agents and combining their results. It covers when to delegate, how to split tasks, and which independent roles can work in parallel.
A project-context summary built from AGENTS.md, related documents, the request, and selected code. AGENTS.md is a file that records project rules and working instructions.
A compact, structured brief for giving a coding task to another agent. It includes the allowed files, relevant requirements, existing tests, limits, commands, and excluded work.
A systematic list of boundary cases and unusual situations a feature may encounter. It covers inputs, failures, timeouts, retries, and simultaneous actions.
An evidence record for completion claims such as “fixed” or “tests passed.” TDD, or test-driven development, is a practice of first seeing a test fail, then making it pass, and checking it again after cleanup.
A requirements review that checks concerns beyond visible features, including performance, maintainability, security, accessibility, and system monitoring. It is designed for a Japanese workflow phase called requirement expansion.
A process for rewriting a reviewed plan into one self-contained final version before approval. It removes review-history notes and keeps only the currently agreed decisions.
A decision-making recipe for deciding whether a code change is ready to become a pull request, based on evidence, review results, risks, checks, and a rollback plan. A pull request is a proposed change that others review before it is added to a codebase.
A repository scan that finds references to files or folders before they are deleted, moved, or renamed. A repository is the project’s stored source files and related documents.
A review that scans initial requirements for missing details, unclear points, hidden assumptions, exceptions, and items that should be excluded. Requirements are the agreed description of what a feature must do.
A workflow for delegating planned coding tasks to separate subagents and checking each result in two reviews: one for matching the specification and one for code quality. A subagent is a separate agent assigned to a focused task.
A team-design skill for deciding when a software task needs multiple agents and assigning roles such as planning, implementation, review, security review, and test review.
★not rated 2 todayA152 tokens
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: