Coding standards, frameworks, refactoring, and code review (16.4K skills)
Design commands and processing loops that reach the same final state after retries or crashes. Use the current state to apply only necessary changes, and make leftover locks, files, or records recognizable during the next run.
Guides additions, refactors, and rewrites by removing unnecessary complexity first. It prioritizes deleting dead code, redundant checks, empty references, and repetitive prompt instructions, then building around observed usage and explicit requirements.
Reviews the current conversation from three perspectives to find reusable lessons and propose concrete skill improvements. It separates findings into accepted, rejected, and backlog items, with user approval required before applying accepted edits.
Review code comments in selected files or changes with a dedicated agent, remove accepted findings, and fix underlying problems. It also proposes ways to enforce documented constraints through types, runtime checks, tests, or automated checks.
Collect review feedback and discussion comments from the pull request for the current branch. Get a grouped summary, a prioritized action list, and unresolved questions that need clarification before you proceed.
Explore a codebase to explain how a feature or subsystem works, how execution flows, and which files or packages own each responsibility. Build understanding before changes and clarify where new code should belong.
Guide debugging from reproducing a failure to finding the first incorrect decision in the code. Distinguish valid states that need handling from invalid states that reveal earlier faults, and look for the same defect elsewhere.
Turn substantial editing, migration, analysis, or verification work into a small, reusable tool. Produce a script, generator, or shared agent skill that performs or checks the work and lets reviewers rerun it.
Explore a codebase with SocratiCode to locate features, functions, and types. Search by meaning, inspect how files depend on each other, and find relevant database schemas or API specifications before reading selected files.
Set up and maintain SocratiCode indexes, which prepare project files for search. Manage file watching, refresh changed content, rebuild dependency graphs, and check infrastructure health when indexing or search results need attention.
Turn a completed workflow or conversation into a reusable agent skill. Clarify inputs, outputs, dependencies, and error handling, then produce an approved design, written instructions or helper scripts, and validation based on the original process.
Prepare the Python package manager uv for skills that depend on it. This skill checks for an existing installation, installs uv if needed, and configures the command search path so uv commands are available.
Check maintainer documentation locally before automated CI checks run. Find missing metadata, outdated review dates, nonexistent paths in the architecture manifest, and completed plans that still contain unfinished tasks, with guidance for correcting common failures.
Check whether a pull request is ready to merge by combining local lint, formatting, and tests with GitHub Actions status and PR description requirements. Returns a readiness verdict with specific reasons for any failed checks.
Fill missing interface translations in a viewer JSON file using English and Chinese source strings. Preview missing entries, translate into six languages through OpenRouter, and validate the updated file with JSON checks and tests.
Compare project documentation with code to find outdated descriptions, missing details, and inconsistent references. Review recent changes, audit all documentation, or focus on one area, then apply approved fixes while preserving the existing language and style.
Process GitHub issues from selection through code fixes and pull requests, which propose changes for review. Parallel agents implement and test fixes, while review monitoring handles follow-up comments. Filters, fork support, and recurring checks help organize the work.
Collect, list, and count actionable comments from CodeRabbit and Codex on GitHub pull requests. Separate comments attached to code from review summaries, check Codex status signals, and manage CodeRabbit review conversations.
Delegate feature development, code refactoring, and pull request reviews to Codex, Claude Code, Pi, or OpenCode. It guides background execution, progress monitoring, and parallel work in separate working directories.
Use the ClawHub command-line tool to find and install agent skills, update them to the latest or a specified version, list installed skills, and publish skill folders with version numbers and change notes.
Generate a Claude CLI prompt from a structured JSON skill idea. The prompt guides creation of a complete skill directory, including SKILL.md, reference documents, helper scripts, and test scaffolding that follows repository conventions.
Extract skill ideas from recent Claude Code session logs using repeated actions, errors, and automation requests as signals. Remove duplicates, score novelty, feasibility, and trading value, then maintain a prioritized backlog for skill generation.
Turn an existing conversation and knowledge of the codebase into a specification, then publish it to the project's issue tracker. It organizes the problem, user stories, implementation and testing decisions, and scope without a new requirements interview.
Turn a plan, specification, or conversation into small development tickets that each deliver a verifiable feature. Identify prerequisite tickets, review the breakdown with the user, and publish approved tickets as local files or tracker issues.