smith-ralph
Ralph Loop integration patterns for iterative AI development. Use when starting Ralph loops, managing iterations, or recovering from context resets. Covers TDD, debugging, context management, and memory persistence.
smith-python
Python development with uv, pytest, ruff, and type hints. Use when writing Python code, running tests, managing Python packages, or working with virtual environments. Covers import organization, type hints, pytest patterns, and environment variables.
smith-prompts
Prompt engineering standards for AI interactions with cache optimization. Use when writing AI prompts, optimizing context usage, or structuring AGENTS.md files. Covers prompt caching, token efficiency, and progressive disclosure patterns.
smith-principles
Fundamental coding principles (DRY, KISS, YAGNI, SOLID, HHH). Use when starting any development task, evaluating implementation approaches, or reviewing code quality. Always active as foundation for all development decisions.
smith-preflight
Pre-ship gate — check the current change against the smith invariants (branch-first, secret-scan, pr-ownership, external-write, verify-before-assert, suggestions, subagent-contract) and report PASS, FAIL, SKIP or N/A (not applicable) per check with a GO or NO-GO verdict. Invoke with /smith-preflight; /smith-ship runs it as step 0 and will not push on NO-GO.
smith-worktree
Claude Code worktree TOOLS — EnterWorktree/ExitWorktree, the bgIsolation guard, worktree.baseRef, the branch-naming gotcha, and the squash-merge sync protocol. Use when invoking EnterWorktree/ExitWorktree, hitting the bgIsolation guard, or cleaning up after a worktree-based PR merge. For raw `git worktree` commands see smith-git.
smith-validation
Hypothesis testing, adversarial verification of findings, root cause analysis, and debugging techniques. Use when debugging, investigating any question whose answer you will report as fact, verifying whether a claim or finding is true, red-teaming a conclusion, testing hypotheses, validating solutions, proving correctness, or performing root cause analysis on failures.
smith-typescript
TypeScript development standards for frontend and backend projects. Use when working with TypeScript, configuring path aliases, setting up test runners (Vitest/Jest), or organizing test files. Covers Vite alias configuration and type checking.
smith-tools
Tool configurations for IDEs, MCP integrations, and development tools. Use when configuring IDE settings, MCP tools, or pytest. Covers conditional tool activation, configuration hierarchy, and synchronization patterns.
smith-tickets
Ticket creation by convention — create Jira issues/sub-tasks in Job Story form, Japanese description, under the correct parent Epic, after reading the relevant context. Invoke with /smith-tickets.
smith-tests
Testing standards and TDD workflow. Use when writing tests, running test suites, implementing TDD, or organizing test files. Covers unit vs integration test separation, pytest patterns, and test-driven development methodology.
smith-subagents
Subagent spawning and return discipline — read-only by default, return findings not actions, treat every return as a claim to verify, reconcile against live state before mutating shared artifacts. Use when spawning Task/Agent subagents, delegating investigation, orchestrating parallel agents, or when a subagent will read or modify shared state (PRs, issues, files, remotes).
smith-review
Multi-round local review loop — review the current worktree change with all relevant smith review skills plus Claude Code review tools (/code-review, /review-pr, CodeRabbit, code-simplifier), iterating until convergence, without shipping. Invoke with /smith-review.
smith-ctx-cursor
Cursor context management with /summarize command, @ mentions for file inclusion, and @codebase discovery. Use when operating in Cursor IDE or when context exceeds 60%. Activate for context optimization in Cursor sessions.
smith-xml
XML tag standards for runtime prompts that mix instructions with embedded data (subagent prompts, assembled system messages). Use when constructing such a prompt, NOT when authoring a SKILL.md body — see smith-skills for that. Covers tag conventions for Claude, GPT-5.x, Gemini, and Harmony, with source-credibility notes verified 2026-07-11.
smith-ctx-kiro
Kiro-specific context management with terminal limitations, Serena MCP as mandatory tool, and file operation workarounds. Use when operating in Kiro IDE. LOAD FIRST in all Kiro sessions - critical platform constraints that prevent hangs and failures.
smith-ctx
Universal context-management foundation — proactive context-level checks and reset recommendations. Use when context usage grows high, when deciding whether to /clear or /compact, or when choosing what to retain across a context reset.
smith-design
SOLID principles and architecture design patterns. Use when starting new features, refactoring code, conducting architecture reviews, or applying object-oriented design principles.
smith-dev
Development workflow standards and code quality requirements. Use when starting ANY task that will modify repo files (implement, fix, add, refactor, harden), initializing projects, running quality checks, or managing agent tasks. Covers branch-first setup, pre-commit checks, task decomposition, and script organization patterns.
smith-dialectic
Socratic interview that stress-tests a plan against project docs and code, one question at a time. Use when user says "grill my plan", "challenge this", "stress-test", or wants to reach shared understanding before implementing.
smith-gh-cli
GitHub CLI gotchas and best practices. Use when executing gh commands. Covers token efficiency, pagination limits, and common pitfalls.
smith-gh-pr
GitHub PR workflows including creation, review cycles, merge strategies, stacked PRs (creation, merge order, rebase after parent merges, squash handling), posting review findings, and confirming a CodeRabbit review actually ran. Use when creating PRs, stacked PRs, or dependent PRs, replying to review comments, running or interpreting a CodeRabbit review (GitHub App or `coderabbit` command line), merging branches, or fetching PR threads. Covers rebase decision trees and AI-generated descriptions. For the stacked shipping pipeline see smith-ship (stacked mode).
smith-auto_mode
Claude Code auto mode classifier — what gets blocked, how to recover from a denial without silently retrying, and where to configure trusted infrastructure. Use when the agent invokes a classifier-sensitive action (push, deploy, external-content duplication), when a prior turn ended in "auto mode classifier" denial, or when the user mentions auto mode / `defaultMode` / `hard_deny`.
smith-automation
Claude Code scheduling primitives — /loop (interval, dynamic, bare), CronCreate/List/Delete, ScheduleWakeup, Monitor, and /schedule (Routines), with a decision matrix and provider/session-scope caveats. Use when setting up a recurring or scheduled task, polling for external state, self-pacing iterations, or the user mentions /loop, /schedule, cron, ScheduleWakeup, or Monitor.
smith-git
Git workflow gotchas and non-obvious practices. Use when performing Git commits, merges, branch management, rebasing, or raw `git worktree` commands. Covers GPG signing, atomic commits, raw worktree patterns, and safety flags. For the Claude Code worktree tools (EnterWorktree/ExitWorktree) see smith-worktree.
smith-browser_mcp
Browser MCP plugin reliability for chrome-devtools-mcp and @playwright/mcp. Take each server's own default browser — both resolve to Chrome's stable channel — and set no browser override at all; launching Vivaldi/Brave/Arc/Opera/Edge is forbidden, though attaching to an already-running one over the Chrome DevTools Protocol (CDP) is the documented escape hatch. Use when invoking chrome-devtools-mcp or Playwright MCP tools, editing .mcp.json / settings.json, triaging browser MCP launch failures, or when a site needs an interactive login the user must complete.
smith-checkpoint
Memory checkpoint — save the current session's durable state into all three memory systems (Serena memory, Basic-Memory note, auto-memory) in their required formats. Invoke with /smith-checkpoint.
smith-analysis
Reasoning frameworks and problem decomposition techniques. Use when planning implementation, evaluating arguments, estimating scope, decomposing complex tasks, or applying first principles thinking.
smith-clarity
Cognitive trap detection and logic fallacy identification. Use when making decisions, evaluating approaches, risk assessment, or detecting faulty reasoning in arguments.
smith-ctx-claude
Claude Code context management with /clear, /compact mechanics, stop hook enforcement at 60%, JSONL state recall, tool-output hygiene, auto memory, and system reminders. Reference dumps (hooks, permission modes, agent features, model routing) live in companion REFERENCE.md. Use when operating in Claude Code IDE or when context exceeds 50%.
subagent-authoring
>-
gh-pr-review
>-
gh-api
>-
skill-authoring
>-
agents-authoring
>-
humanizer
>-
linear-cli
>-
python-scripting
>-
command-authoring
>-
hunk-staging
>-
ddd-guide
Document-Driven Development workflow for existing codebases. Provides systematic planning, documentation-first design, and implementation verification.
module-development
Guide for creating new Amplifier modules including protocol implementation, entry points, mount functions, and testing patterns. Use when creating new modules or understanding module architecture.
amplifier-philosophy
Amplifier design philosophy using Linux kernel metaphor. Covers mechanism vs policy, module architecture, event-driven design, and kernel principles. Use when designing new modules or making architectural decisions.
agents-md-generator
Automatically generate comprehensive AGENTS.md files that teach AI agents how to work with a repository's structure, conventions, and guidelines
caps-format-validator
Validate CAPS (Coding Agent Playbook Spec) playbooks for correctness, completeness, and compatibility with Claude Code and Goose
kubernetes-health-check
Verify Kubernetes cluster health, resource availability, and readiness before deploying applications
thrivve-mc-how-many
Thrivve Partners Monte Carlo simulation to forecast story/task completion based on historical throughput. Use when the user asks "how many stories/tasks will be completed by [date]" with historical daily throughput data. Requires at least 10 days of throughput history and a future target date. Provides probabilistic forecasts at specified confidence levels (default 85%).
thrivve-mc-when
Thrivve Partners Monte Carlo simulation to forecast completion date based on remaining work and historical throughput. Use when the user asks "when will I complete [N] stories/tasks" with historical daily throughput data. Requires at least 10 days of throughput history, a count of remaining items, and optional confidence level (default 85%).
speckit
依照 speckit 的標準工作流程，引導式地產生規格文件、澄清需求、產生實作計劃與任務清單，每個階段都需要人工審核確認
learning-graph-generator
Generates a comprehensive learning graph from a course description, including 200 concepts with dependencies, taxonomy categorization, and quality validation reports. Use this when the user wants to create a structured knowledge graph for educational content.
Page 1439 of 1708 · 85390 results
