Goblin1024
34 verified skills102 total stars
cancel
Stop active workflow and clean up state
tools3
review
Reviewer-only pass for /plan --review and cleanup artifact review
tools3
worker
Kimi-native team worker protocol (ACK, mailbox, task lifecycle)
data-ai3
wiki
Persistent markdown project wiki stored under .omk/wiki with keyword search and lifecycle capture
documentation3
skills/build-fix
# Skill: Build Fix This skill triggers when the user encounters compilation, build, or type errors. ## Role You are the **Debugger** and **Executor**. Load `prompts/debugger.md` and `prompts/executor.md` for your operating principles. ## Workflow 1. Run the build/typecheck command to see the errors. 2. Identify the root cause of the error. 3. Make the necessary code modifications to fix the error. 4. Re-run the build command to verify the fix. 5. If the build still fails, repeat steps 2-4. 6.
development3
ai-slop-cleaner
Run an anti-slop cleanup/refactor/deslop workflow
development3
skills/analyze
# Skill: Analyze This skill triggers when the user wants to investigate or deeply analyze a problem or codebase. ## Role You are the **Explorer** and **Debugger**. Load `prompts/explorer.md` and `prompts/debugger.md` for your operating principles. ## Workflow 1. Use semantic search and file system tools to locate relevant context. 2. Trace the control flow and data flow. 3. Identify patterns, bottlenecks, or root causes of issues. 4. Provide a detailed summary of your findings to the user. 5.
tools3
autopilot
Full autonomous execution from idea to working code
development3
autoresearch
Stateful validator-gated research loop with native-hook persistence
research3
skills/code-review
# Skill: Code Review This skill triggers when the user wants a code review. ## Role You are the **Code Reviewer** and **Security Reviewer**. Load `prompts/code-reviewer.md` and `prompts/security-reviewer.md` for your operating principles. ## Workflow 1. Examine the provided code or the recent diff. 2. Check for correctness, readability, performance, and best practices. 3. Identify any security vulnerabilities. 4. Provide a structured review report with specific, actionable feedback. 5. Offer
development3
configure-notifications
Configure OMX notifications - unified entry point for all platforms
tools3
deep-interview
Socratic requirements gathering through structured questioning
tools3
deepsearch
Thorough codebase search
development3
doctor
Diagnose and fix oh-my-kimi installation issues
documentation3
ecomode
Token-efficient execution modifier with budget-aware routing
testing3
frontend-ui-ux
Designer-developer for UI/UX work
development3
git-master
Git expert for atomic commits, rebasing, and history management
tools3
skills/help
# Skill: Help This skill triggers when the user asks for help or wants to see available commands. ## Workflow 1. Display the contents of the Quick Commands section from `AGENTS.md`. 2. Briefly explain the purpose of Oh-My-Kimi (orchestrating Kimi CLI with roles and workflows). 3. Ask the user what they would like to do next.
tools3
hud
Show or configure the OMX HUD (two-layer statusline)
tools3
skills/note
# Skill: Note This skill triggers when the user wants to save a note or finding. ## Workflow 1. Extract the core information from the user's prompt. 2. Save the information into a markdown file in `.omk/context/notes/` with a descriptive timestamped filename. 3. Confirm to the user that the note has been saved.
development3
skills/plan
# Skill: Plan This skill triggers when the user wants to start a planning workflow. ## Role You are the **Planner**. Load `prompts/planner.md` for your operating principles. ## Workflow 1. Analyze the user's request to understand the complex goal. 2. Break down the goal into actionable, sequential tasks. 3. Determine dependencies. 4. Output a structured plan in `.omk/plans/`. 5. Propose the first task to begin implementation.
development3
ralph
Persistence loop until task completion with verification
tools3
ralph-init
Initialize a PRD (Product Requirements Document) for structured ralph-loop execution
documentation3
ralplan
Architecture planning and plan approval workflow
tools3
security-review
Run a comprehensive security review on code
development3
skill
Manage local skills - list, add, remove, search, edit, setup wizard
tools3
swarm
N coordinated agents on shared task list (compatibility facade over team)
data-ai3
tdd
Test-Driven Development enforcement skill - write tests first, always
development3
skills/team
--- name: team description: Parallel agent execution via Kimi-native multi-worker runtime trigger: $team flags: - name: --count description: Number of workers to spawn (default: 2) phases: - starting - dispatching - monitoring - consolidating - completing gates: - type: prompt_specificity description: Task must describe parallelizable work blocking: true - type: proper_decomposition description: Task should indicate sub-tasks for parallel execution blocking: f
testing3
trace
Show agent flow trace timeline and summary
data-ai3
ultraqa
QA cycling workflow - test, verify, fix, repeat until goal met
tools3
ultrawork
Parallel execution engine for high-throughput task completion
tools3
visual-verdict
Structured visual QA verdict for screenshot-to-reference comparisons
testing3
web-clone
URL-driven website cloning with visual + functional verification
development3