「graph」の検索結果

Split a multi-call LM workflow by cognitive load, not by accuracy: let one strong model make the few reasoning decisions and a cheap model do the many mechanical executions (Aider architect+editor, DSPy optimizer-LM vs task-LM, vLLM speculative draft+target, LangGraph supervisor+worker are the same shape). Use when designing or cost-optimizing a pipeline that calls an LM many times, when deciding which steps need a strong reasoner vs a cheap executor, or when adding an escalation valve for when the cheap tier degrades. Search keywords: reduce LLM cost, cheaper model, lower token cost, model cascade, route to cheap model, strong model plus cheap model, LLM cost optimization.

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。

Symbol-level code context for LLM coder-agents: tree-sitter extracts symbols, PageRank ranks them over the cross-file reference graph, and the top class/function signatures are fed to the LLM as a token-budgeted read-only map (not RAG, no vector index, human-auditable). Use when an agent must locate the right files in a large/multi-file repo, when builds/refreshes/scopes a repo-map, or when "model edits the wrong file" needs fixing.

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。

Decision protocol for wiring a verify-then-fix loop around a code-editing LLM agent. The agent edits → runs lint/test → reads the output → fixes → re-runs, bounded by an iteration cap and an escalation rule. Activates whenever a coder agent has a verifiable success criterion (exit code, type-checker output, failing assertion) and the user wants the agent to converge to "green" on its own. Framework-agnostic — wraps Aider's `--auto-lint`/`--auto-test`, an OpenHands SWE-Bench loop, a manual LangGraph cycle, or Claude Code's bash tool just the same.

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。

Enhancement-overlay decision protocol for STREAMING the output of long-running LLM / agent runs from the *backend*, not just wiring a typing animation in the UI. Activates when a coder agent must stream final tokens to a chat client, surface intermediate agent steps (which tool, which node, partial reasoning), emit custom tool-progress events, choose a transport (SSE vs WebSocket), or decide what to do when the client disconnects mid-stream. The langchain / langgraph skills mention stream modes but stop at "you can stream"; this skill encodes *what to stream, over what transport, and how to fail safely*.

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。

Decision protocol for making side-effectful agent tools idempotent — so when an LLM tool call is retried (timeout, framework resume, user re-run, model duplicate emit), the second call is a no-op instead of a double-send. The load-bearing premise: the LM cannot promise it'll call exactly once; the tool must promise the second call is safe. Framework-agnostic — applies to LangGraph node bodies that re-run on resume, MCP tools, OpenAI tool-calling retries, CrewAI delegated tool invocations, and direct HTTP wrappers. Search keywords: duplicate email sent, charged twice, exactly-once, idempotency key, tool called twice, retry side effect, double-send, at-least-once delivery.

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。

Decision protocol for the map-reduce / dynamic fan-out pattern in LM pipelines — "given list L, run f(item) for each item in parallel, then combine". Activates when the coder agent is about to process N items with N LM calls (per-doc summarize, per-query retrieve, per-candidate rank, parallel tool fan-out). Encodes the *when*, *how many at once*, *what to do when one fails*, and *how to reduce* — not the API of any single framework. Cross-framework: LangGraph `Send`, CrewAI parallel tasks / Flow, `asyncio.gather`, `ThreadPoolExecutor`, LlamaIndex batch retrieval.

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。

Enhancement-overlay skill — the DECISION + WIRING layer for LM observability that the single-backend skills [[langsmith]], [[phoenix]], [[mlflow]] do NOT cover. Each of those installs one backend; none of them help you DECIDE which backend fits your stack/scale/budget, nor give you a one-line autolog that turns it on fast. Use when starting any LM project, before the first deploy, or the moment someone asks "why did it do that?" and there are no traces to answer with. The skill picks a backend by stack (LangSmith for LangChain/LangGraph; Phoenix for OSS/local OpenTelemetry; MLflow for ML-shops already on MLflow; Langfuse for self-host), wires one-line autolog, verifies traces land, and adds eval hooks — instrumenting BEFORE you need it. Cross-links the first-debug-move skill [[agentsop-prompt-history-inspect]]. Do NOT activate to re-teach a backend you already chose (defer to its own skill), or for non-LM ML experiment tracking with no LLM calls (that is plain MLflow).

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。

Decision protocol for wrapping a REST / GraphQL / RPC API as a tool an LLM agent can call. The load-bearing premise: the *tool surface* is an LM-friendly subset of the *API surface* — one tool per user intent, not one per endpoint. Activates when a coder agent must expose an external HTTP API to a model (function calling, tool_use, MCP, LangChain `@tool`, CrewAI `BaseTool`). Encodes the *what to surface, how to name, how to shape, how to fail* — not any single framework's API. ~80% of agent tools in production are HTTP wrappers; this is the SOP for getting them right.

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。

Tool skill — the *first move* in any LM-debugging session: dump the actual rendered prompt the framework sent to the model, before changing anything else. Activate when an LM call produced an unexpected output (wrong answer, schema violation, refusal, truncation, cost spike, latency spike, infinite loop, "model got dumber after upgrade"). The skill enforces a 30-second inspect step BEFORE any prompt edit, model swap, retry, or temperature tweak. Cross-framework cheat sheet: DSPy `inspect_history`, LangGraph `get_state_history`, CrewAI `step_callback`, LangChain `set_debug`/`set_verbose`, Aider `/diff`+`--verbose`, raw OpenAI/Anthropic via `OPENAI_LOG=debug`/`ANTHROPIC_LOG=debug` or HTTPX event hooks. Do NOT activate for first-time prompt authoring, exploratory prompt design, or non-LM bugs.

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。

Review a complete mathematical-modeling competition paper with an atomic checklist covering abstract readability, page structure, typography, headings, pagination, figures, tables, equations, anonymity, metadata, visual balance, and page allocation. Return the review in chat or Markdown by default, and provide the bundled single-sheet checklist when the user requests a reusable form. Use when the user provides a PDF or Word paper and asks for a full-paper format and writing-compliance audit; do not replace technical model correctness review.

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。

Check a mathematical-modeling paper's in-text citations, bibliography, appendix length and content, code, supporting files, reproducibility, consistency, and anonymity. Use when the user provides a paper or related materials and wants one consolidated P0-P3 audit; do not generate or invent missing sources or programs.

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。

Audit application source code against the OWASP Top 10 (2021) vulnerability categories — broken access control, cryptographic failures, injection, insecure design, security misconfiguration, vulnerable components, authentication failures, data integrity, logging failures, SSRF. Use when the user mentions 'OWASP,' 'OWASP Top 10,' 'security audit,' 'security review,' 'secure code review,' 'code security review,' 'vulnerability audit,' 'find vulnerabilities,' 'appsec review,' 'application security audit,' 'check for security issues,' 'broken access control,' 'IDOR,' 'SQL injection,' 'XSS,' 'SSRF,' or wants to check their codebase for common security weaknesses.

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。

Digital archiving with AI enrichment and entity extraction. Use when building content archives or knowledge graphs.

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。

Draft, confirm, and post replies to GitHub PR review threads. Handles per-category reply formatting, re-fetches thread resolution state so auto-resolved threads are skipped, posts via GraphQL, and resolves threads whose entry carries a resolution reason. Use when the user asks to "reply to PR threads", "post PR thread replies", or "draft PR reply messages".

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 for social science academic paper drafting, outlining, diagnosis, review, and revision. Trigger for topic diagnosis, research question refinement, literature review planning, CNKI/Google Scholar literature search-to-Zotero workflows, structured paper reading notes, theoretical framework design, research design critique, draft review, section or paragraph revision, citation and evidence risk checking, causal-claim checking, and pre-submission review. Do not use for purely creative writing or non-academic editing.

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 production-ready Node.js backend services with Express/Fastify, implementing middleware patterns, error handling, authentication, database integration, and API design best practices. Use when creating Node.js servers, REST APIs, GraphQL backends, or microservices architectures.

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。

Database schema patterns for InsForge including social graphs, e-commerce, content publishing, and multi-tenancy with RLS policies. Use when designing data models with relationships, foreign keys, or Row Level Security.

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。

Fetch unresolved PR review comments using GitHub GraphQL API, filtering out resolved and outdated feedback.

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。

Generate or edit images using AI models (FLUX, Gemini). Use for scientific illustrations, diagrams, schematics, infographics, concept visualizations, and artistic images. Supports image editing to modify existing images (change colors, add/remove elements, style transfer). Useful for figures, posters, and visual explanations.

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。

Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.

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。

Master REST and GraphQL API design principles to build intuitive, scalable, and maintainable APIs that delight developers. Use when designing new APIs, reviewing API specifications, or establishing API design standards.

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。

Track complex, multi-session work with dependency graphs using bd (beads) issue tracker. Use when work spans multiple sessions, has complex dependencies, or requires persistent context across compaction cycles. For simple single-session linear tasks, TodoWrite remains appropriate.

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 content in any language indistinguishable from a skilled human writer. Avoids AI-sounding phrases, varies sentence structure, creates natural prose. Detects target language and loads language-specific AI-tells, register, and orthography from `languages/`.

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。

Complete UI/UX design capability — a wiki-style knowledge map of design theory (UX laws, Nielsen heuristics, usability, visual hierarchy, typography, contrast/WCAG, design systems, aesthetics-vs-conversion) fused with a hands-on execution playbook for building distinctive, production-grade frontend interfaces. Use when designing or building any UI (websites, landing pages, dashboards, components, apps), reviewing or critiquing UI/UX, diagnosing why a design doesn't convert, or styling/beautifying web interfaces. Grounds every visual decision in UX principles and avoids generic AI aesthetics.

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。