gosu-mcp-core

Install and configure the Gosu MCP server

Install gosu-mcp-core skills to Codex CLI user directory (~/.codex/skills). Skills are reusable bundles with SKILL.md instructions and optional scripts/resources. Usage: `/install-gosu-skills-to-codex [skill-name]` or `/install-gosu-skills-to-codex` to install all.

Install skill scripts to /usr/local/bin using symlinks

Execute all remaining tasks defined in a markdown task file using task-list-md skill

Use this agent only when you are instructed. Examples: <example>Context: User wants to invoke subagent "dynamic" . user: invoke subagent dynamic with this prompt "use system prompt `code-reviewer` to review this file ..." assistant: 'I'll invoke the dynamic agent with your provided prompt'</example> <example>Context: User wants to use code-reviewer prompt via dynamic agent. user: use dynamic subagent with prompt ID "code-reviewer" to review this module pkg/otel assistant: 'I'll use the dynamic subagent with the "code-reviewer" prompt to review the pkg/otel module.'</example>

This agent MUST BE USED when you need to review Python, Go or TypeScript source code or test files for best practices, code quality, and completeness. Examples: <example>Context: The user request you to implement a new service class and wants you to review it before committing. user: 'Implement the UserService class in src/services/user_service.py then do a code review' assistant: 'I'll use the gosu-code-reviewer agent to perform a comprehensive review of this UserService after i finish its implementation.' <commentary>Since the user wants code review, after finish the implementation use the Task tool to launch the gosu-code-reviewer agent to analyze the Python/TypeScript code for best practices and quality.</commentary></example> <example>Context: The user has written tests and wants to ensure they follow best practices and have good coverage. user: 'Please review my test file user.service.test.ts to make sure it follows testing best practices' assistant: 'Let me use the gosu-code-reviewer agent to analyze your TypeScript test file for best practices and completeness.' <commentary>The user is requesting test file review, so use the gosu-code-reviewer agent to examine and review the test file.</commentary></example> <example>Context: The user has implemented a new handler in Go and wants a code review. user: 'Please review my handler.go file for idiomatic Go and best practices.' assistant: 'I'll use the gosu-code-reviewer agent to analyze your Go handler for idiomatic usage and best practices.' <commentary>The user is requesting Go code review, so use the gosu-code-reviewer agent to examine and review the Go source file.</commentary></example>

This agent MUST BE USED when you need to review Markdown specification files (*.spec.md) for completeness, accuracy, and alignment with project requirements. Examples: <example>Context: User has just created or updated a specification file for a new API module. user: 'Review the user-authentication.spec.md file i've just finished writing for our new auth module' assistant: 'Let me use the gosu-spec-reviewer agent to review your specification file for completeness and alignment with the design requirements' <commentary>Since the user has created/updated a spec file, use the gosu-spec-reviewer agent to analyze it against the required structure and quality standards.</commentary></example> <example>Context: The user request you to update a spec file and wants you to review it before committing. user: 'Update src/application-name-api/controllers/data.controller.spec.md to add validation for all data fields then do a spec review' assistant: 'I'll use the gosu-spec-reviewer agent to perform a comprehensive review of this spec file after i finish updating it.' <commentary>Since the user wants spec review, after finish updating the spec file use the Task tool to launch the gosu-spec-reviewer agent to analyze the Markdown spec file for best practices and quality.</commentary></example>

Use this agent when you need to verify that source code implementations match their corresponding specification files and clean up fully implemented spec details. Examples: <example>Context: User has written a new function and wants to verify it matches the spec requirements. user: 'I just implemented the authentication middleware in auth.go, can you check if it matches auth.spec.md?' assistant: 'I'll use the gosu-spec-verifier agent to compare your auth.go implementation against the auth.spec.md specification and verify compliance.' <commentary>Since the user wants to verify implementation against spec, use the gosu-spec-verifier agent to analyze both files and report on implementation status.</commentary></example> <example>Context: User wants to clean up spec files after completing implementation work. user: 'I've finished implementing all the features in user_manager.go, please update the associated spec file to remove completed details' assistant: 'I'll use the gosu-spec-verifier agent to verify your user_manager.go implementation and clean up the user_manager.spec.md file by removing fully implemented details.' <commentary>Since the user wants spec cleanup after implementation, use the gosu-spec-verifier agent to verify and clean the spec file.</commentary></example> <example>Context: User wants to clean up all spec files in a directory after a large implementation push. user: 'Please clean up all *.spec.md files in the pkg/gosu/ directory to remove fully implemented details.' assistant: 'I'll use the gosu-spec-verifier agent to verify all implementations in pkg/gosu/ and update each spec file by removing details for items that are fully implemented.' <commentary>Since the user wants to clean up all spec files in a directory, use the gosu-spec-verifier agent to process each *.spec.md file in the directory and update them accordingly.</commentary></example>

This agent MUST BE USED when you need to execute test scripts, test suites, test files, run integration tests, end to end (e2e) tests, and then analyze failed test cases to understand what went wrong. Examples: <example>Context: User has written new code and wants to verify all unit test cases are still pass. user: 'I just updated the authentication service, can you run the tests and check if anything broke?' assistant: 'I'll use the gosu-test-runner agent to run the test suite and analyze any failures.' <commentary>Since the user wants to run tests and identify failures, use the gosu-test-runner agent to execute tests and analyze any failed cases.</commentary></example> <example>Context: User is debugging failing integration tests. user: 'The integration API test cases in `test/integration` directory are failing, can you run the tests and tell me what's going wrong?' assistant: 'Let me use the gosu-test-runner agent to execute the tests in `test/integration` and identify the specific failures causing the API test cases to fail.' <commentary>User needs test execution and failure analysis, so use the gosu-test-runner agent.</commentary></example> <example>Context: User wants to run all e2e tests. user: 'Can you run all e2e tests and summarize the results?' assistant: 'I'll use the gosu-test-runner agent to execute all e2e tests and provide a summary of the results.' <commentary>Explicitly running e2e tests and summarizing results is a task for the gosu-test-runner agent.</commentary></example>

Utility tool to create new git worktrees with proper setup for development environments. Automates creation, copying, symlinking, and ownership management for smooth developer onboarding or isolated task/feature work.

Utility scripts for GitHub pull request management. Includes tools for fetching bot-generated review comments (linters, security scanners, dependabot), replying to review threads programmatically, listing merged pull requests with filtering, resolving conversations, and automating PR review workflows. Useful for batch processing comments, CI/CD integration, quality metrics tracking, release notes generation, and automated responses to bot reviewers.

This skill should be used when the user asks to "create a session hook", "add a temporary hook", "create a hook for this session", "set up a hook that activates immediately", "configure a one-time hook", or mentions "session-specific hook", "per-session hook", or "temporary hook". Provides guidance for creating session-scoped hooks that activate immediately without restarting Claude Code.

Parse and manage tasks list in a markdown file using checklist format (tasks.md, checklist.md)

13 hooks across 11 events