Fantasy player comparisons need rankings that match the sport, scoring format, and time period. This skill retrieves published Ball Ranks data and cites its source.

Video prompts can produce weak motion, drifting subjects, or clips with the wrong duration. This skill guides Layer model selection, image anchoring, motion prompts, and loop checks.

Inconsistent datasets and incompatible models can weaken a trained style while still consuming Creative Units. This skill guides curation, training choices, weight tuning, and contribution checks before generation.

Generated pixel art can have blurred edges, inconsistent palettes, and mismatched animation frames. This skill guides Layer workflows using explicit grids, approved references, and sprite sheet tools.

Vague visual briefs lead to inconsistent assets and repeated generation costs. This skill guides style decisions, moodboard comparisons, palette constraints, and storyboard planning before production.

Unspecified topology, materials, and reference inputs can lead to unsuitable meshes and extra paid runs. This skill guides staged Layer workflows with model-specific settings and price checks.

Creative teams need to coordinate models, reference files, and costs across asset generation tasks. This skill guides Layer MCP requests from workspace selection to completed assets and style warnings.

Unclear goals and changing priorities can make agent decisions diverge from your expectations. Align captures confirmed criteria in ALIGN.md and reviews meaningful changes against them.

Web animations can drop frames when rendering costs or main-thread work exceed the frame budget. This skill identifies likely causes and recommends focused, verifiable improvements.

CPU hotspots and memory allocation pressure can make deployed Workers and Durable Objects difficult to diagnose. This skill guides profiling, evidence analysis, and targeted optimization.

OT teams need clear technical scope without unsupported authority or pricing conclusions. This skill builds validated DOCX descriptions and a separate cost-analysis handoff.

Other Transaction budgets can mix cost shares, milestone payments, fees, and unlike price bases. This skill builds traceable workbooks with neutral comparisons and validated formulas.

Federal acquisition research can lose traceability across documents, sources, calculations, and decisions. This skill organizes a staged workflow with approvals, evidence IDs, and validated reports.

Federal buyers need transparent LH and T&M estimates that separate labor, travel, materials, and ceilings. This skill builds validated workbooks from documented assumptions and government benchmarks.

Federal FFP estimates require traceable wages, indirect costs, benchmarks, travel, and formulas. This skill builds a validated workbook with visible assumptions and sourced calculations.

Cost-reimbursement estimates require traceable wages, indirect pools, fee treatment, travel, and benchmark data. This skill builds auditable CPFF, CPAF, and CPIF workbooks with validation steps.

Federal contracting research is fragmented across sources and often lacks clear evidence boundaries. This skill structures approved research, records sources, and produces traceable findings or validated briefs.

Federal acquisition policy changes across codified rules, deviations, notices, and guidance. This skill produces dated, sourced analysis with explicit status and decision boundaries.

Lovable 项目需要谨慎协调 GitHub、Supabase 和部署。此技能为 Claude、Codex 和 Claude Code 提供结构化提示词和工作流规则，用于 Lovable 后端变更。

Lovable 后端部署可能需要反复进行手动提示和检查。此技能可通过 MCP 或浏览器自动化来自动化边缘函数和迁移工作流，并报告验证结果。