290 件 ・ 関連度順
概要と使いどころ
Unified LLM torch-profiler triage skill for `sglang`, `vllm`, `TensorRT-LLM`, and `TokenSpeed`. Use it to inspect an existing `trace.json(.gz)` or profile directory, or to drive live profiling against a running server when supported and return one three-table report with kernel, overlap-opportunity, and fuse-pattern tables.
日本語の概要は準備中です。原文の説明を表示しています。
Use when the user has written (or wants to write) a `ModularPipelineBlocks` subclass in a local Python file and needs to package it into a Hub-uploadable directory. Covers the workflow from a single `block.py` file to a published custom-block repo that consumers can load via `ModularPipeline.from_pretrained(<repo>, trust_remote_code=True)`.
日本語の概要は準備中です。原文の説明を表示しています。
Detect and prevent dependency confusion (public-over-private package name resolution) in npm, PyPI, and Maven by enumerating claimable internal package names with tools like `confused` and OWASP `dep-scan`, then enforcing source restrictions via `.npmrc`, `pip.conf`/`pyproject.toml`, and Maven `settings.xml`. Use when onboarding a repo to a supply-chain security program, auditing lockfiles/manifests for confusable dependencies, or after an incident that may have leaked internal package names.
日本語の概要は準備中です。原文の説明を表示しています。
Drive LinkedIn Sales Navigator and Recruiter through the Curviate CLI. Covers `sales-nav` (search, profile, message, lead and account lists, save-lead, save-account) and `recruiter` (projects, pipeline, talent-search, save-candidate, project jobs, applicants, message, profile, search), and what can refuse them: the LinkedIn subscription on the connected account, the Curviate seat, and the beta-operations gate. These commands carry no Curviate add-on paywall. Use when working a Sales Navigator or Recruiter seat, or when diagnosing why a premium command failed.
日本語の概要は準備中です。原文の説明を表示しています。
Quality gate via second model. Spawn a different AI model to review work before committing. Includes refusal routing: if one model refuses, switch silently to the next. Extended in v0.25.1 with structured review-mode gating (when to invoke vs not) and a Codex code-review handoff for the diff-review case.
日本語の概要は準備中です。原文の説明を表示しています。
Universal archivist for personal file archives (Dropbox/B2/Gmail-takeout/local-mount/hard-drive-dump). Filters for high-value content (the user's own writing, ideas, relationships) and surfaces it interactively. REFUSES TO RUN without an explicit gbrain.yml `archive-crawler.scan_paths:` allow-list.
日本語の概要は準備中です。原文の説明を表示しています。
Converts long-form markdown (specs, RFCs, reports, plans, explainers) into a single-file, lightly-interactive HTML document with sticky TOC, scrollspy, search filter, code-copy buttons, and design-system-driven brand tokens. Triggers when the markdown-html-orchestrator classifies an input as DOCUMENT, or when invoked directly via /cs:md-document. Reads the design-system config via config_loader.py and inlines the user's 12 derived CSS custom properties; refuses to render if onboarding hasn't run. Single-file output — Google Fonts + Prism.js CDN are the only externals; no framework runtime, no build step. Use after orchestrator routing or after design-system onboarding is confirmed.
日本語の概要は準備中です。原文の説明を表示しています。
Captures the user's brand identity once via a 10-question onboarding wizard (primary/accent HEX + heading + body Google Fonts + design style editorial/technical/minimal/playful + default output directory + syntax theme + TOC behavior + optional logo/company), validates body-text and link contrast against WCAG 2.2 AA, derives 12 CSS custom properties in HSL space, and stores the result for every markdown-html converter to consume. Use before any markdown-html conversion. Triggers on first-run onboarding ("set up the brand", "configure markdown-html", "run onboarding"), on explicit reset ("reset the design system", "re-onboard"), and is checked by every converter via config_loader.py before rendering. Refuses to save if body-text contrast fails AA 4.5:1 or the output dir isn't writable. Precedence is project (./.markdown-html/) > global (~/.config/markdown-html/) > built-in defaults; MARKDOWN_HTML_NO_CONFIG=1 bypasses.
日本語の概要は準備中です。原文の説明を表示しています。
Converts a markdown deck (slides separated by `---` HR boundaries or by `# ` H1 headings, with optional `<!-- notes: ... -->` presenter notes blocks) into a single-file HTML presentation with arrow-key / space / PgDn / PgUp / Home / End / P / Esc keyboard navigation, presenter mode (split view with current slide + speaker notes + clock + next-slide preview), URL-hash deep linking, and `@media print` page-per-slide for PDF export. Triggers when the markdown-html-orchestrator classifies an input as SLIDES, or when invoked directly via /cs:md-slides. Reuses md-document's markdown parser for slide-body rendering and reads design-system tokens via config_loader.py. Refuses if input has no clear slide boundaries, produces a 1-slide deck, or `--strict-notes` is on with < 50% notes coverage. Use after orchestrator routing.
日本語の概要は準備中です。原文の説明を表示しています。
Use when a user wants to convert any markdown file in their Claude project into a single-file, lightly-interactive HTML — long-form documents (specs, plans, RFCs, reports, explainers), code reviews with diffs and severity-tagged annotations, or slide decks. Triggers on "convert this markdown to HTML", "make this an HTML file", "turn this into an interactive document", "render this report as HTML", "PR writeup as HTML", "slides from this markdown". Forks context to route to one of three converter sub-skills (md-document, md-review, md-slides) based on a deterministic doctype classifier, after the user has run the design-system onboarding once. Refuses if input is under 100 lines (per Shihipar — markdown still wins below the threshold) or design-system isn't onboarded. Distinct from Anthropic's official Playground plugin (which is interactive prompt-tuning controls with sliders/knobs/prompt-copy-back) and from marketing/landing/ (which is a landing-page generator).
日本語の概要は準備中です。原文の説明を表示しています。
Use when running an annual SaaS audit, doing category-level spend review, or rationalizing the supplier base — when the user needs a spend audit, spend categorization (UNSPSC-aligned with Pareto breakdown and industry profiles), purchasing-cycle analysis (bottleneck categories per Goldratt's Theory of Constraints), or risk-balanced supplier consolidation that refuses single-source recommendations for tier-1 categories without a documented break-glass plan. Triggers on "spend audit", "SaaS audit", "spend categorization", "supplier rationalization", "supplier consolidation", "category strategy", "duplicate SaaS", "renewal cluster".
日本語の概要は準備中です。原文の説明を表示しています。
Retune a skill corpus for a new model, measurement-first: mine the run archive for a baseline, establish a noise floor, audit the corpus adversarially, then cut in measured passes until a pre-registered bar clears. Requires a benchmark harness that can A/B two builds of the corpus; refuses without one.
日本語の概要は準備中です。原文の説明を表示しています。
Make the desktop's model provider fail on demand (connection refused, DNS not found, socket reset, 429, 5xx, 401, 402, stalled or cut stream) with a local fault server and test providers. Use when reproducing or designing model error and retry states without a real outage.
日本語の概要は準備中です。原文の説明を表示しています。
Use when an n8n-mcp account targets more than one n8n instance — i.e. the `n8n_instances` tool is available, the user mentions multiple n8n instances or environments (prod vs staging, several teams or clients), a workflow / datatable / credential / execution call returns an unexpected NOT_FOUND or reads data you don't recognize, or a credential create/update/delete is refused with an `INSTANCE_AMBIGUOUS` error. Covers choosing and switching which instance this MCP session targets, verifying the target before high-stakes work — credential writes above all — and recovering from misroutes and ambiguous-write fail-closes. Always consult this skill before operating on a specific instance, before any credential create/update/delete on a multi-instance account, or when a call hits the wrong/empty data or an `INSTANCE_AMBIGUOUS` error.
日本語の概要は準備中です。原文の説明を表示しています。
Write JavaScript or Python for the n8n Custom Code Tool (@n8n/n8n-nodes-langchain.toolCode) — the AI-agent-callable tool, NOT the workflow Code node. Use when building a Code Tool attached to an AI Agent, writing code that an LLM will invoke, parsing the `query` input, returning a string result, defining an input schema for structured arguments (specifyInputSchema, jsonSchemaExample, DynamicStructuredTool), or troubleshooting errors like "Wrong output type returned", "No execution data available", "The response property should be a string, but it is an object", "Cannot assign to read only property 'name'", or an AI agent that refuses to call the tool. Covers the critical differences between Code node and Code Tool: return format (string vs `[{json:{...}}]`), unavailability of `$fromAI`/`$input`/`$helpers` in the Code Tool sandbox, naming rules for AI invocation, and when to use `toolWorkflow`/HTTP Request Tool instead.
日本語の概要は準備中です。原文の説明を表示しています。
Stores, retrieves, and manages data as objects in Cloud Storage on Google Cloud (also known colloquially as GCS) buckets. Use when you need to interact with Cloud Storage — set up a Storage MCP server (remote or local Toolbox), create or configure buckets, upload, download, stream, or transfer data, organize objects with folders, generate signed URLs, control access (IAM, ACLs, public access prevention), set storage classes (Standard, Nearline, Coldline, Archive), manage lifecycle and cost, protect data (versioning, CMEK, retention, Bucket Lock, holds, soft delete), host static websites, trigger Pub/Sub notifications, mount buckets (gcsfuse), or optimize performance. Covers gcloud storage / gsutil, JSON/XML APIs, client libraries, Terraform, and Cloud Storage MCP servers. Don't use for non-Storage MCP servers, block storage (Persistent Disk), BigQuery, or databases (Cloud SQL, Spanner, Bigtable, Firestore).
日本語の概要は準備中です。原文の説明を表示しています。
Diagnoses GKE persistent-storage failures — volume attach/mount errors (Regional PD on optimized VMs, fsGroup mount timeouts), disk-performance and node storage-pressure issues, slow-disk Pod-creation failures, volume-expansion problems, Local SSD / Hyperdisk Storage Pool creation errors, and Cloud Storage FUSE OOM. Use when Pods are stuck in ContainerCreating, volumes fail to attach or mount, or nodes report storage pressure. Don't use for routine storage provisioning or StorageClass/PVC authoring (see the gke-storage skill).
日本語の概要は準備中です。原文の説明を表示しています。
Build and publish a Gradio demo on Hugging Face Spaces for a user-provided LoRA. Use when someone asks to create, generate, ship, or publish a Space, demo, Gradio app, or playground for a LoRA — including LoRAs for Qwen-Image, Qwen-Image-Edit, LTX-Video, Wan, FLUX, SDXL, or other diffusion base models. Also triggers when someone describes a LoRA they trained or hosts on the Hub and wants to share it. Covers picking the right base pipeline and `diffusers` inference recipe, designing a UI tailored to the LoRA's task and inputs (Union/multi-task control, edit, video, image, etc.), respecting model-card recommendations (trigger words, steps, guidance, LoRA scale, example inputs), and shipping to ZeroGPU hardware as a private Space by default.
日本語の概要は準備中です。原文の説明を表示しています。
Uninstall a community skill that was installed via the hub. Confirms before deleting files, refuses to touch first-party plugin skills, and logs every action. Use when the user wants to fully remove a community skill ("uninstall [skill]", "remove this skill") rather than just disable it.
日本語の概要は準備中です。原文の説明を表示しています。
Reference: detailed uninstall, disable, and re-enable workflows for community skills installed via the legal builder hub. Safe by default — refuses to touch first-party plugin skills, confirms before removing files, and logs every action. Loaded by the /legal-builder-hub:uninstall and /legal-builder-hub:disable skills.
日本語の概要は準備中です。原文の説明を表示しています。
Prove a Convex feature works — seed, drive as multiple mocked users via convex-test, assert behavior including the negative authz cases (wrong user refused, data-scope enforced).
日本語の概要は準備中です。原文の説明を表示しています。
Comprehensive offensive methodology for software supply chain attacks covering the full kill chain from reconnaissance through exploitation. Addresses dependency confusion across npm, PyPI, and NuGet ecosystems where internal registry override allows an attacker to inject malicious packages that shadow private dependencies. Covers typosquatting techniques for popular packages, compromised package injection via maintainer account takeover or social engineering, and build system attacks through Makefile injection, setup.py install hooks, and npm postinstall scripts. Extends into CI/CD artifact tampering where build outputs are replaced or modified in transit, code signing abuse through stolen or self-signed certificates, upstream repository compromise via commit injection or force-push to trusted repos, and container image supply chain attacks including base image trojaning and registry confusion. Maps to MITRE ATT&CK T1195.001 (Supply Chain Compromise: Compromise Software Dependencies and Development Tools) and T1195.002 (Supply Chain Compromise: Compromise Software Supply Chain). Integrates tooling such as confused for dependency confusion scanning and dependency-check for known vulnerable component detection. Each technique section provides reproducible proof-of-concept patterns, detection guidance for defenders, and engagement-safe execution notes for authorized red team operations.
日本語の概要は準備中です。原文の説明を表示しています。
Create a gooey blob system using SVG filters where multiple shapes merge into a single fluid form. Use overlapping circles combined with a Gaussian blur and color matrix filter to produce a continuous, organic mass. The forms should visually fuse and separate based on proximity. Focus on filter-driven merging (blur + threshold effect), soft organic boundaries with no hard edges, multiple independent shapes behaving as one system, and smooth continuous motion that feels fluid and cohesive.
日本語の概要は準備中です。原文の説明を表示しています。
Write JavaScript or Python for the n8n Custom Code Tool (@n8n/n8n-nodes-langchain.toolCode) — the AI-agent-callable tool, NOT the workflow Code node. Use when building a Code Tool attached to an AI Agent, writing code that an LLM will invoke, parsing the `query` input, returning a string result, defining an input schema for structured arguments (specifyInputSchema, jsonSchemaExample, DynamicStructuredTool), or troubleshooting errors like "Wrong output type returned", "No execution data available", "The response property should be a string, but it is an object", "Cannot assign to read only property 'name'", or an AI agent that refuses to call the tool. Covers the critical differences between Code node and Code Tool: return format (string vs `[{json:{...}}]`), unavailability of `$fromAI`/`$input`/`$helpers` in the Code Tool sandbox, naming rules for AI invocation, and when to use `toolWorkflow`/HTTP Request Tool instead.
日本語の概要は準備中です。原文の説明を表示しています。