sentimony · Agent Skills
出典を確認した実質的な文書を持つ Skill が 61 件あります。カードを開くと原文、パッケージ内容、導入方法を確認できます。
implement
Task-by-task execution of an existing plan in the current session, with plan drift handling and verification. Use when executing a plan directly, including resuming interrupted execution. Not for subagent-driven execution (subagents-implement).
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
maintain-agents-md
Create, audit and maintain repository agent instructions and context architecture. Use when working on AGENTS.md, CLAUDE.md, .claude/rules/, or instruction-related SKILL.md content - always-loaded context cost, stale, duplicated or conflicting instructions, agent memory drift. Not for human-facing docs or skill authoring (skill-crafting).
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
md-preview
Render saved Markdown, JSON, YAML, TOML, CSV or text files, including requests, specs and plans, as HTML previews with a table of contents, cross-file links, and a maintained folder README index. Use after an authoring workflow saves or edits a file, or when the user explicitly requests a preview. Not for arbitrary Markdown editing.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
negafix
Bans negative parallelism, the "it's not just X, it's Y" construction. Use when writing or substantively editing project prose (docs, READMEs, marketing copy), or when asked to audit, score, or clean it up. Not for ordinary negation.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
parallel-agents
Independence proof, state isolation, wave width and result reconciliation for concurrent agent work. Use when several units of work might run concurrently through agents - independent investigations, specialist reviews, repository analyses, or plan tasks that look unrelated.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
plan-craft
Multi-step implementation plan from an approved design or settled requirements. Use when a plan is needed before code changes.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
plan-crafting
Multi-step implementation plan from an approved design or settled requirements. Use when a plan is needed before code changes.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
plan-execution
Task-by-task execution of an existing plan in the current session, with plan drift handling and verification. Use when executing a plan directly, including resuming interrupted execution. Not for subagent-driven execution (plan-orchestration).
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
plan-orchestration
Plan execution through coordinated subagents - delegation, review, verification, integration. Use when executing an existing plan via subagents, including resuming interrupted orchestration. Not for inline execution (plan-execution) or unrelated parallel investigations.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
prose-craft
Edits prose to fit its reader. Use when drafting, rewriting, editing, or reviewing prose in any language for clarity, concision, tone, or audience fit (docs, messages, articles, product copy). Not for translation, agent instructions, UI design, dash audits (dashfix), negative-parallelism audits (negafix), or review findings (received-review).
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
prose-crafting
Edits prose to fit its reader. Use when drafting, rewriting, editing, or reviewing prose in any language for clarity, concision, tone, or audience fit (docs, messages, articles, product copy). Not for translation, agent instructions, UI design, dash audits (dashfix), negative-parallelism audits (negafix), or review findings (review-resolution).
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
review-request
Independent code review of an implementation against its requirements and the Git diff. Use when a completed or partially committed implementation, committed or in the working tree, needs that review.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
review-resolution
Technical validation of review feedback before any fix. Use when code-review findings, PR comments, CI review output, or reviewer suggestions (human, subagent, GitHub, static analysis) need a verdict or disposition.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
scope-triage
Classify a request and settle design before code. Use when a request needs design decisions - new features, product or UX behavior, architecture changes, unclear success criteria, or two materially different approaches. Specified mechanical refactors, localized fixes with known behavior, and single-outcome config changes go straight to implementation.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
session-learnings
Records lasting learnings from the session in the project's gitignored docs/session-learnings/AGENTS.md without asking, and on a separate request triages the notes into AGENTS.md, CLAUDE.md, global instructions, or a stronger enforcement such as a check or test, applying only approved edits. Use when pitfalls, user corrections, conventions, or commands that finally worked should reach future agents, whatever the conversation language, and when an installed agentflow chain calls it after finalize
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
skill-crafting
Decide whether a recurring workflow needs a reusable agent skill, and design, evaluate, improve, split, merge or retire skills. Use when working on trigger boundaries, architecture, behavior contracts or effectiveness evaluation. Not for agent instruction files (maintain-agents-md), settled planning (plan-craft) or code fixes (tdd).
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
subagents-implement
Plan execution through coordinated subagents - delegation, review, verification, integration. Use when executing an existing plan via subagents, including resuming interrupted orchestration. Not for inline execution (implement) or unrelated parallel investigations.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
tdd
Red-green-refactor development with meaningful failing tests first. Use when implementing testable behavior changes - new features, bug fixes or refactors with observable contracts. Not for purely mechanical edits without behavior changes.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
transfer-context
Save and resume work across agent sessions with a verifiable checkpoint. Use when the user asks to save, resume or check a checkpoint, or to hand work to a new session.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
typescript
TypeScript-specific problem solving and type-safe API design. Use when working on compiler errors, tsconfig, strict typing, inference, generics, complex types, module resolution, ESM/CJS interop, type-check performance, framework type-check issues, improving type safety, migration or monorepo config. Not for routine feature work where types are incidental.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
verification-gate
Evidence-backed pass or fail verdict on a completion claim. Use when work is about to be called done, fixed, ready, or mergeable, before a merge or pull request, a branch finish, or a handoff report.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
verify-gate
Evidence-backed pass or fail verdict on a completion claim. Use when work is about to be called done, fixed, ready, or mergeable, before a merge or pull request, a branch finish, or a handoff report.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月16日
vitest
Vitest setup, test writing, debugging and migration in JavaScript/TypeScript projects. Use when configuring, writing, running, migrating or auditing Vitest tests - Vite, Vue, Nuxt, React, Next.js, Node libraries, workspaces, coverage, mocks, snapshots, flaky tests, CI parity or Jest migration.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日
web-debug
Python Playwright testing of local web apps. Use when verifying frontend functionality, debugging UI behavior, capturing browser screenshots, or viewing browser console logs.
- 1時間の導入数
- 3
- GitHub Stars
- 8
- 更新日
- 11月13日