shimo4228
@shimo4228
Public Skills
skill-stocktake
by shimo4228
"Use when auditing Claude skills and commands for quality. Supports Quick Scan (changed skills only) and Full Stocktake modes with sequential subagent batch evaluation."
ai-era-architecture-principles
by shimo4228
"Framework adoption decision matrix: custom vs large frameworks in the Claude Code era. Use when evaluating whether to adopt a large framework or build custom with AI."
algorithm-migration-with-rollback
by shimo4228
"Use when replacing a core algorithm (encryption, hashing, ML model) that affects persisted user data and needs rollback safety."
cjk-aware-text-metrics
by shimo4228
"CJK/Latin weighted token estimation for multilingual LLM pipelines. Use when processing Japanese/Chinese/Korean text with fixed chars-per-token constants."
backward-compatible-frozen-extension
by shimo4228
"Use when extending a frozen dataclass or Pydantic pipeline with new fields without breaking existing consumers."
claude-code-mcp-manual-install
by shimo4228
"Use when adding MCP servers from within a Claude Code session where CLI is unavailable. jq-based ~/.claude.json editing workaround."
brainstorming-communication
by shimo4228
"Use when the user is exploring ideas and hasn't decided on a direction yet. Provide information without forcing choices."
long-document-llm-pipeline
by shimo4228
"Use when processing documents over 50K characters through LLM APIs with section splitting and batch cost reduction."
deep-research-api-landscape
by shimo4228
"Use when building automated research systems. Official Deep Research APIs (OpenAI, Gemini, Perplexity) over browser automation."
skill-stocktaking-process
by shimo4228
"Use when learned skills exceed 10 per location or need consolidation. 4-step stocktaking with character budget awareness."
root-cause-challenge-pattern
by shimo4228
"Use when evaluating whether a new feature, dependency, or abstraction is truly needed. 5-step root cause challenge."
json-data-validation-test-design
by shimo4228
"Use when validating auto-generated JSON data files too large for manual review. Layered pytest validation pattern."
python-optional-dependencies
by shimo4228
"Use when supporting multiple backends/providers without forcing all dependencies on users. pyproject.toml extras pattern."
python-immutable-accumulator
by shimo4228
"Use when building immutable state accumulators in Python. Frozen dataclass + tuple pattern with slots gotcha."
xcode-package-swift-misidentification
by shimo4228
"Use when an iOS app builds successfully but the simulator does not launch due to Package.swift misidentification."
prh-hyphen-regex-escape
by shimo4228
"Use when adding hyphenated terms to prh spelling dictionaries on Node.js 20+. Unicode regex flag incompatibility."
swift-actor-persistence
by shimo4228
"Use when building a thread-safe data persistence layer in Swift using actors with in-memory cache and file storage."
claude-code-self-generation-over-api
by shimo4228
"Use when Claude Code needs to generate or transform text data in bulk. Try self-generation before API calls on Max plan."
xcode-pbxproj-file-registration
by shimo4228
"Use when xcodebuild fails with 'cannot find in scope' after adding a .swift file outside Xcode. Register in 4 pbxproj sections."
content-hash-cache-pattern
by shimo4228
"Use when caching expensive file processing results. SHA-256 content-hash keying with frozen CacheEntry and service layer wrapper."
mock-friendly-api-layering
by shimo4228
"Use when mock assertions fail due to public functions forwarding internal params like url or timeout to a shared helper."
service-layer-extraction
by shimo4228
"Use when a Typer/Click CLI module exceeds 300 LOC with mixed concerns. Extract business logic into a testable service layer."
zenn-context-driven-writing
by shimo4228
"Use when preparing context and drafts for experience-based Zenn articles. Parallel research + context file pattern."
directory-structure-enforcement-hooks
by shimo4228
"Use when Claude Code creates files in wrong locations despite CLAUDE.md rules. PreToolUse + Stop hook 3-layer defense."
swift-codable-decode-diagnosis
by shimo4228
"Use when debugging Swift Codable JSON decode errors with vague localizedDescription messages."
zenn-markdownlint-config
by shimo4228
"Use when setting up markdownlint-cli2 for a Zenn content repository. Zenn-specific rule overrides for false positives."
cost-aware-llm-pipeline
by shimo4228
"Use when building an LLM-powered app that needs cost control via model routing, budget tracking, retry, and prompt caching."
zenn-qiita-crosspost-workflow
by shimo4228
"Use when cross-posting a Zenn article to Qiita. Automated conversion pipeline with platform-specific adaptations."
regex-vs-llm-structured-text
by shimo4228
"Use when parsing structured text (quizzes, forms, documents). Start with regex, add LLM only for low-confidence edge cases."
tech-writing-patterns
by shimo4228
"Use when writing or reviewing technical articles for Zenn/Qiita. Cross-posting, tone adjustment, quality patterns."
python-module-to-package-refactor
by shimo4228
"Use when splitting a Python module (>400 LOC) into a package. mock.patch target update rules and checklist."
keyword-based-llm-eval
by shimo4228
"Use when evaluating LLM-generated structured output against expected results using keyword matching and F1 metrics."
protocol-di-testing
by shimo4228
"Use when testing Swift code that depends on file system, network, or external APIs via protocol-based dependency injection."
zenn-textlint-workarounds
by shimo4228
"Use when textlint produces false positives on Zenn articles. prh hyphen loop, :::message, and --fix workarounds."
claude-code-tool-patterns
by shimo4228
"Use when writing large files (>1000 lines), editing after many operations, or configuring hooks in Claude Code."
textual-tui-pipeline-interception
by shimo4228
"Use when adding an interactive review/approval step to a CLI pipeline using Textual TUI with immutable state."
immutable-model-updates
by shimo4228
"Use when designing Swift model structs that need predictable, thread-safe state updates without mutation."
cross-source-fact-verification
by shimo4228
"Use when fact-checking drafts containing dates, quantities, or causal claims against multiple independent sources."
parallel-subagent-batch-merge
by shimo4228
"Use when generating 50+ structured items with parallel Claude Code subagents and merging outputs into one file."
data-generation-quality-metrics-loop
by shimo4228
"Use when validating auto-generated data quality through iterative generate-measure-fix loops with quantitative metrics."