メニュー
SkillsMP は bsamiee/Maghz から 24 件の skill を収集しています。skill を開くとソースと詳細を確認できます。
収集済み skill 24 件中 24 件を表示しています。
Owns all prose, markdown files and the comments inside source files and fences through a register law, an anti-anchoring rule against rosters, counts, and freezes, a named defect catalog, file-kind templates, and a deterministic gate and fixer. Use when…
原文の言語: 英語
Owns work placement and delegation craft across Claude Code's execution surfaces — main turn, `/btw`, fork, subagent, background task, nested subagent, agent team, dynamic workflow — with the selection economics that pick one, the delegation contract that…
原文の言語: 英語
Bounded one-shot Gemini sub-calls through the Antigravity CLI (`agy`): multimodal judgment over screenshots, generated images, UI states, and diagrams, read-only adversarial review lanes that return typed falsifiable findings as a second-model perspective,…
原文の言語: 英語
Builds and hardens AppleScript, JXA, and Open Scripting Architecture automation — osascript/osacompile runners, Apple Events and object-specifier dispatch, TCC consent preflight and hardened-runtime entitlements, OSAKit/NSAppleScript embedding,…
原文の言語: 英語
Local code review through three engines — CodeRabbit (working tree), Greptile (committed commits vs base), Macroscope (in-place AST correctness) — driven by one verb rail (`launch`, `status --follow`, `kill`, `findings --normalize`, `slice`, `reconcile`,…
原文の言語: 英語
Dispatch self-contained work to Codex (gpt-5.6 Sol/Terra) through the `codex` MCP tool or the `codex` CLI — an isolated agent returning one report; writes, design, hard reviews, fleet fix waves, research, and volume extraction all ride it. Trigger on repo…
原文の言語: 英語
Use for Bash 5.2+/5.3 scripts, shell CLIs, entrypoints, CI jobs, cron, text-processing pipelines, ShellCheck remediation, and shell reviews — "write a shell script", "fix this bash", "make a CLI wrapper". Enforces strict mode, dispatch-table routing,…
原文の言語: 英語
Use for PostgreSQL 18 SQL, migrations, DDL, functions, RLS policies, indexes, query plans, extensions, and SQL embedded in TypeScript via @effect/sql-pg. Enforces set-algebraic queries, schema-level invariants, current PostgreSQL features, extension-first…
原文の言語: 英語
Current library, framework, and API documentation through the Context7 MCP tools `resolve-library-id` and `query-docs` — signatures, configuration, setup, and code examples from live indexed docs instead of training-data recall. Use for setup or configuration…
原文の言語: 英語
Generates and validates GitHub Actions workflows and custom actions (composite/Docker/JavaScript) with SHA-pinned supply chain security, SLSA attestation, OIDC federation, and harden-runner enforcement. Use when creating, editing, reviewing, or validating…
原文の言語: 英語
Owns the persistent behavior surfaces of the Claude Code harness and the placement law that picks one: the CLAUDE.md hierarchy, imports, path-scoped rules, the memory content law and scored audit, auto memory, settings scopes and precedence, permission…
原文の言語: 英語
Interact with Heptabase using the CLI to manage knowledge base content, search cards, edit properties, read parsed PDF and media transcript content, export local files, manage whiteboard cards, and browse AI Tutor goals, courses, and lessons.
原文の言語: 英語
Builds Claude Code and Codex hooks as one discipline — command, HTTP, MCP-tool, prompt, and agent handlers across the full event surface. Use when building PreToolUse validation, PostToolUse formatting or redaction, PermissionRequest auto-approval,…
原文の言語: 英語
Hostinger estate operations across the whole account API: the domain portfolio (registration, DNS zone records, nameservers, forwarding, WHOIS, lock and privacy), the VPS fleet (lifecycle, Docker Compose projects, firewalls, SSH keys, snapshots, recovery,…
原文の言語: 英語
Author single-file, self-contained interactive HTML artifacts — reports, dashboards, roadmaps, plans, decision docs, and their variants (diff-reviews, quizzes, wargames, buy-in packets, boards, decks, explainers, atlases) — as typed deliverable compositions…
原文の言語: 英語
Adversarial elicitation that converts unknowns into durable decision artifacts: corpus-grounded interviews whose questions confront the user with what exists, blindspot passes along a named interrogation-axis catalog, direction brainstorms and wargames,…
原文の言語: 英語
Generates and validates Mermaid diagrams with YAML frontmatter, ELK layout, Dracula theme tokens, and a bundled render, graph-logic, and SVG-geometry validator. Owns diagram methodology — when to diagram, node and edge selection, per-type construction,…
原文の言語: 英語
Owns the autonomous PR lifecycle over `gh` and the GitHub GraphQL API — from landed work to a squash-merged PR with no branch remaining: every reviewer (CodeRabbit, Greptile, Macroscope, Codex, Claude, human) awaited on its true completion signal with…
原文の言語: 英語
Use this skill for any task that creates, modifies, inspects, or destroys cloud infrastructure or SaaS configuration, from one-off CLI operations to full multi-resource projects, across providers in the Pulumi ecosystem — AWS, Azure, GCP, Kubernetes,…
原文の言語: 英語
Drives the McNeel Rhino-MCP-Platform `mcp__rhino-mcp-platform__*` tools for interactive Rhino host work: slot lifecycle, run_csharp/run_python scripting, headless document IO, scene query/selection, viewport capture, and GH2 Grasshopper graph authoring. Use…
原文の言語: 英語
Estate secret custody and consumption over 1Password and Doppler — op as the permanent local store, Doppler as the runtime backend, scoped CLI reads, directory-scope resolution, the SessionStart pull rail, service-token custody classes, secret-file…
原文の言語: 英語
Designs, authors, reviews, and repairs skill bundles — SKILL.md roots, trigger descriptions, references, templates, bundled scripts, frontmatter policy — and proves them with trigger and adherence evals. Use when creating a skill, rewriting or auditing an…
原文の言語: 英語
Web research through the Tavily CLI as one skill — search (ranked fact-finding with a server-side synthesized answer so raw pages never enter context), research (a cited multi-source report), extract (clean markdown from up to 20 known URLs), map (a site's…
原文の言語: 英語
Author runnable workflow scripts for Claude Code's Workflow tool — deterministic multi-agent orchestration files under `.claude/workflows/` that fan out fresh-context subagents through plain JavaScript control flow: the `meta` block, `agent()`, `pipeline()`,…
原文の言語: 英語