spread1000-builder

Formulates an AI for Science research plan. Proposes how to integrate AI into the researcher's domain through web research and literature survey, and generates a concrete 研究計画書 (research plan document). Use when user doesn't know how to apply AI to their 研究テーマ (research theme), wants to create a 研究プラン (research plan), or wants to learn about AI for Science latest trends (最新動向).

Support post-award research management and reporting for SPReAD. Guide grant application procedures, interim progress review (3 months), final outcome report (6 months), accounting report preparation, budget reallocation between expense categories, research plan change approval requests, publication acknowledgment templates, and research data management implementation. Use when you want to know post-award procedures, write an interim report, create a final report, change budget allocations, reallocate between expense categories, change research plans, or learn how to write publication acknowledgments.

Draft the SPReAD proposal (research plan document). Integrate the research plan, Azure architecture design, and cost estimate to generate a proposal aligned with the 公募要領 requirements. Use when writing the proposal, finalizing the research plan document, or preparing submission documents.

Estimate Azure costs based on architecture design. Generate monthly and full research period (~180 days) cost estimates and produce cost breakdowns required for SPReAD budget applications. Use when estimating Azure costs, planning budgets, or filling in expense sections of the proposal.

Designs a research infrastructure architecture on Microsoft Azure based on the research plan. Generates a comprehensive design document including GPU clusters, ML pipelines, data storage, and network configuration. References Microsoft Learn MCP for optimal configurations. Use when user wants to design Azure構成 (Azure architecture), build GPUクラスタ (GPU cluster), or design MLパイプライン (ML pipeline).

Generate step-by-step experiment procedure documents for running AI research on deployed Azure infrastructure. Covers environment setup, data preparation, training execution, inference, result evaluation, and checkpoint/reproducibility management. Use when you want to create an experiment procedure guide, run experiments on Azure ML, execute GPU training jobs, set up experiment environments, or manage reproducibility. 実験手順書を作りたい、Azure上で実験を実行したい、学習ジョブを走らせたい、 実験環境をセットアップしたい、再現性を管理したい場合に使用。

Guide the step-by-step Azure deployment procedure for SPReAD research infrastructure. Covers prerequisite checks, OIDC federation setup, Bicep deployment execution, post-deployment verification, and rollback procedures. Use when you want to deploy Azure resources, set up OIDC authentication, execute Bicep deployment, verify deployed resources, or troubleshoot deployment errors.

Collects missing context from the user via one-question-at-a-time dialogue, generates a 6-element meta-prompt, and passes it to downstream skills. Applies shikigami's interactive purpose-discovery pattern specialized for SPReAD. Use when user request is ambiguous (依頼が曖昧), research field is unknown (研究分野が不明), or required information is insufficient (必要情報が不足).

Generate Azure research infrastructure system architecture diagrams, data flow diagrams, and network diagrams in draw.io format using the draw.io MCP Server (simonkurtz-MSFT). 700+ official Azure icons are built in and can be retrieved via fuzzy search with search-shapes. Automatically creates visual architecture diagrams from the Phase 1 (azure-architect) design document. Use when you want to create system architecture diagrams, draw.io architecture diagrams, Azure configuration diagrams, or data flow diagrams.

Perform a comprehensive final review of the SPReAD proposal before submission. Cross-validate all phase deliverables (research plan, Azure architecture, cost estimate, proposal, diagrams) for consistency, score against the 6 peer review criteria with actionable improvement suggestions, validate character limits and mandatory items, and generate a submission-readiness report. Use when performing a final check before submission, validating cross-document consistency, running a mock peer review, or checking submission readiness.

Generate Bicep templates and GitHub Actions CI/CD pipelines based on the Azure configuration design document. Manage research infrastructure as code and enable reproducible deployments. Use when you want to generate Bicep templates, build CI/CD, or set up IaC.

Guide the entire SPReAD submission process. Covers AI interview preparation and execution, e-Rad registration and submission procedures, eligibility and overlap restriction checks, and completeness verification of submission documents (様式0–4). Use when preparing for the AI interview, learning e-Rad procedures, checking eligibility, checking overlap restrictions, learning student application procedures, or assembling submission documents.

Optimize Agent Skill description fields for maximum routing precision and discovery accuracy. Use when a skill is not being activated correctly, descriptions overlap between skills, or improving Tool Coverage score.

Capture learnings, edge cases, and pitfalls from completed tasks and maintain Gotchas sections in Agent Skills for Memory Persistence. Use when finishing a task, resolving a bug, completing a review, or when an agent made a mistake that should be prevented in future sessions.

Audit Agent Skills and their environment against the Harness 7-axis framework. Evaluates AGENTS.md orchestration, SKILL.md quality, assets/references usage, MCP configuration, and Custom Agent design. Use when evaluating skill quality, scoring harness maturity, diagnosing why a skill underperforms, or running a pre-release harness review.

Design AGENTS.md Orchestrator configurations with WHEN/DO routing patterns, phase gates, urgency triage, and multi-skill workflow coordination. Use when setting up a new multi-skill workflow, designing phase transitions, creating task classification trees, or generating AGENTS.md files for suites.

Discover the true objective for Agent Skills development through structured one-question-at-a-time dialogue. Extracts requirements, structures prompts, and generates optimized specifications before skill generation begins. Use when a user's skill development request is ambiguous, lacks detail, or needs clarification before generating Agent Skills packages.

Generate Harness-optimized Agent Skill packages with full suite support. Creates AGENTS.md orchestrators, SKILL.md sub-skills, Custom Agents, copilot-instructions.md, .mcp.json, and assets/references/scripts directories. Use when creating a new skill suite from scratch, bootstrapping a skill folder, or generating any component of a Harness-optimized skill package.