「search run」の検索結果

Run a Meta-Harness-style optimization loop NATIVELY — automatically search over the scaffolding around a FIXED base model (memory, retrieval, context construction, prompt templates, summarization, tool-selection logic) by proposing candidate variants, scoring each on a cheap deterministic eval, and keeping a Pareto frontier of quality vs cost — using native Agent / Workflow / loop tools instead of a standalone Python harness. Use this whenever the user wants to optimize, evolve, tune, distill, or search over a harness, scaffold, prompt system, memory or retrieval policy, context-assembly code, or summarizer while keeping the model fixed; whenever they mention Meta-Harness, harness optimization, scaffold evolution, automatic prompt/memory optimization, an evolutionary or Pareto search over candidate implementations, or "make the harness/agent better without retraining"; and whenever the gain must come from the code AROUND the model rather than the model weights. Reproduces the Meta-Harness paper's method natively, with no claude_wrapper.py and no metered solver API.

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。

Orchestrate work as a cyclic directed graph of subagent-executed nodes with transition criteria on edges, budgeted cycles, and per-node gates (runtime verification, metric, or artifact). Use when the user says "workgraph", "run this as a graph", "graph this work", or when a goal has parallel branches, feedback loops, or mixed gate types that a linear loop cannot express. Every node is sized to a 64k context budget, so small goals run as a two-node do/verify graph and large goals scale out without caps. For clock-driven or metric-driven single loops, timeboxed-iterating and autoresearch remain lighter.

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。

Curate the project's knowledge base and memory in .unikit/memory/ — the indexed reference library of coding standards and framework usage patterns. ALWAYS use this skill whenever the user adds knowledge from a source: a URL, a file, a folder path, a PDF, an article, or a book — any path/link/document is a strong signal for the knowledge base, not a quick rule. Also use it to research or document how the project uses a framework, or to add a researched core/stack knowledge entry. Sub-commands by phrasing: "migrate the rules" or "promote rules" runs migrate-rules; "optimize the knowledge base" or "optimize the rules" runs optimise (extract large sections into references); "validate" re-syncs RULES_INDEX.md; "--module <id>" targets a module. For a quick one-line convention or correction typed as a prompt ("always do X", "never use Y", "remember this") with no source, use /unikit-rules; for architecture decisions use ARCHITECTURE.md.

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。

Outbound SDR (Sales Development Representative) operating system that runs the full top-of-funnel motion across HubSpot (CRM, engagement, and source of truth) and Exa (web research + signals). Use this skill WHENEVER the user is doing outbound sales development, building target/prospect lists, defining or refining an ICP, enriching leads, researching accounts or people before outreach, writing cold emails or multi-touch sequences, qualifying or scoring leads, logging activity to CRM, handing meetings off to AEs, or reporting on pipeline. Trigger it even when the user only says things like "find me 50 leads", "research this account", "write a cold email to this person", "score these leads", "log this in HubSpot", or "build a sequence"; these are all SDR tasks this skill governs. Do not wait for the user to say "SDR" or "prospecting" explicitly.

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。

Scientific research engine with adversarial review, tree search, and serendipity detection. Use when: exploring hypotheses, validating findings against literature, running computational experiments with quality gates, or hunting for unexpected discoveries. Do NOT use for simple Q&A, code editing, or non-research tasks.

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。

Autonomous goal loop — give it ONE goal plus how to verify it, the agent runs check→fix→check itself and comes back only with the finished result (code with audit, research with fact-check, content with critique, audit/backtest). Self-verification instead of constant supervision, with hard safety guards. Vary the loop with a verification rubric designed before it runs. Triggers — "work autonomously to the result", "don't come back until it's done", "run the loop yourself", "drive it to done".

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。

Run remote BLAST searches against NCBI servers using Biopython Bio.Blast.NCBIWWW. Use when identifying unknown sequences, finding homologs, picking the correct BLAST program (blastn/blastp/blastx/tblastn/tblastx/psiblast/megablast/dc-megablast), interpreting Karlin-Altschul E-values, avoiding the max_target_seqs trap (Shah 2019), choosing composition-based statistics, or limiting searches by organism. Covers RID lifecycle, database choice (nt/nr/refseq_select/swissprot), word-size and CBS taxonomy.

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。

Observe the user's screen via screenpipe, detect repeated research workflows, match them against existing scientific-agent-skills, and draft new skills (or composition recipes that chain existing ones) for the patterns not yet covered. Use when the user asks to analyze their recent work and propose skills based on what they actually do. Requires the screenpipe daemon (https://github.com/screenpipe/screenpipe) running locally on port 3030 — the skill has no other data source and will refuse to run if screenpipe is unreachable. All detection runs locally; only redacted cluster summaries reach the LLM.

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。

Autonomous experiment loop that optimizes any file by a measurable metric. Inspired by Karpathy's autoresearch. The agent edits a target file, runs a fixed evaluation, keeps improvements (git commit), discards failures (git reset), and loops indefinitely. Use when: user wants to optimize code speed, reduce bundle/image size, improve test pass rate, optimize prompts, improve content quality (headlines, copy, CTR), or run any measurable improvement loop. Requires: a target file, an evaluation command that outputs a metric, and a git repo.

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。

Research one agent harness's release (docs, source, runtime) and fill its capability cells in scripts/harness_capabilities.py with cited evidence. Use after a harness version bump, when just harness-probe reports re-research, or when capability cells read unknown.

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。

Run 150+ AI apps via inference.sh CLI - image generation, video creation, LLMs, search, 3D, Twitter automation. Models: FLUX, Veo, Gemini, Grok, Claude, Seedance, OmniHuman, Tavily, Exa, OpenRouter, and many more. Use when running AI apps, generating images/videos, calling LLMs, web search...

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。

Full SolanaOS operator skill — install the runtime, query Solana blockchain data via SolanaTracker RPC, manage wallets, run OODA trading loops, configure multi-venue strategies (spot + Hyperliquid + Aster perps), control BitAxe mining fleets, pair Seeker devices, and interact with the SolanaOS Hub. Use when asked about SolanaOS installation, Solana wallet balances, token research, trading strategy, OODA loops, gateway setup, Telegram bot configuration, mining fleet management, or any SolanaOS runtime operation.

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。

Work the review of a research run with rs_review — read the manuscript and record the panel's concerns with the evidence that grounds each one, keep the author's reply and the re-review's finding in their separate columns, run the devil's advocate with the concession threshold the tool enforces, and write the panel report and traceability matrix a response-to-reviewers letter is built from. Use when a document is ready for review, when a concern has to be answered or re-reviewed, or when an attack or a concession is being recorded.

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。

Hand a research run over as a file — read `rs_open action=capabilities` for the formats this build writes, choose between `md`, `docx`, a self-contained `html`, the `ris`/`bib`/`enw` library exports and `tex`, pick `draft=watermark` while working or `draft=clean` to hand it over, and read `build/build.json` for what was built from which bytes. Use when a document is finished enough to leave the run, when a citation style or an export has to be chosen, or when a build refuses.

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。

Run autonomous research tasks that plan, search, read, and synthesize information into comprehensive reports.

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。

Run autonomous research tasks that plan, search, read, and synthesize information into comprehensive reports.

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。

Product-launch command desk (Busabase App-in-Skill) built on the RAMP discipline — Research / Assemble / Mobilize / Prove. The agent assembles the launch checklist and drafts launch assets, channel submissions (Product Hunt / Hacker News / directories), press pitches, and the launch-day runbook; the human approves the launch-readiness gate and steers launch day in a review UI. Use when the user invokes $kelly-launch or /kelly-launch, or mentions launching / shipping a product, a launch checklist or runbook, launch readiness, RAMP, Product Hunt / Show HN / press-kit / launch-email drafting, or wants to review and approve launch assets and channel submissions before they go public. This is launch OPERATIONS — the checklist, assets, submissions, and runbook — not a launch video (for a promo video use product-launch-video).

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。

Plan and execute reproducible computational research with explicit inputs, environments, provenance, deterministic steps, validation, and workspace outputs. Use for analyses that must be rerun, audited, or handed to another researcher.

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 AI scientist systems with the ToolUniverse Python SDK for scientific research. Covers the 3 calling patterns (`tu.run` portable dict API, `tu.tools.X` function API, direct class instantiation), tool loading, batch execution, MCP server integration, and embedding-based tool search. Use for SDK programming, custom tool composition, benchmarking pipelines, and integrating ToolUniverse into research workflows.

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。

Windows アプリ開発 (Windows AI Foundry) の AI/ML 統合リファレンス。 Windows AI APIs、LanguageModel, TextRecognizer, ImageScaler, ImageObjectExtractor, ImageObjectRemover, ImageDescriptionGenerator, SemanticSearch, ContentModeration, Phi Silica LoRA、Windows ML、ONNX Runtime、execution providers、LearningModel、 Foundry Local、CLI、REST API、SDK、model catalog、MCP サーバー、App Actions、 IActionProvider、Agent Launchers、DirectML、IDMLDevice、operators、tensors。

Run remote BLAST searches against NCBI servers using Biopython Bio.Blast.NCBIWWW. Use when identifying unknown sequences, finding homologs, picking the correct BLAST program (blastn/blastp/blastx/tblastn/tblastx/psiblast/megablast/dc-megablast), interpreting Karlin-Altschul E-values, avoiding the max_target_seqs trap (Shah 2019), choosing composition-based statistics, or limiting searches by organism. Covers RID lifecycle, database choice (nt/nr/refseq_select/swissprot), word-size and CBS taxonomy.

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。

Interact with DeerFlow AI agent platform via its HTTP API. Use this skill when the user wants to send messages or questions to DeerFlow for research/analysis, start a DeerFlow conversation thread, check DeerFlow status or health, list available models/skills/agents in DeerFlow, manage DeerFlow memory, upload files to DeerFlow threads, or delegate complex research tasks to DeerFlow. Also use when the user mentions deerflow, deer flow, or wants to run a deep research task that DeerFlow can handle.

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。

Run remote BLAST searches against NCBI servers using Biopython Bio.Blast.NCBIWWW. Use when identifying unknown sequences, finding homologs, picking the correct BLAST program (blastn/blastp/blastx/tblastn/tblastx/psiblast/megablast/dc-megablast), interpreting Karlin-Altschul E-values, avoiding the max_target_seqs trap (Shah 2019), choosing composition-based statistics, or limiting searches by organism. Covers RID lifecycle, database choice (nt/nr/refseq_select/swissprot), word-size and CBS taxonomy.

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。

[21] UNDERSTAND. Deep research workflow for this project using 150% scope (100% core + 50% boundary), evidence-based reasoning, and structured investigation notes. Use when the task requires investigation, root-cause analysis, or mapping unknown areas. Always maintain a research log file that captures findings, hypotheses, and next branches; use web.run when external verification is needed.

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。