tak

FDE archetype pack for integration and vendor-platform engagements: contract design, adapters against undocumented interfaces, and COTS configuration.

3 skills

Prompt & context engineering team — agents (prompt-architect, prompt-implementation-engineer, prompt-reliability-engineer) that own the layer between the model and the app: prompt & context strategy (task decomposition, pattern selection across few-shot / CoT / decomposition / role, system-vs-user structure), the output-format contract (JSON mode, tool/function-calling schemas, few-shot curation, token budgeting), and prompt reliability (eval sets & regression, prompt versioning/CI, prompt-injection & jailbreak defense at the prompt layer). Provider-neutral. skills, a knowledge bank with Mermaid decision trees + a dated 2026 reference, best-practices, templates. Seams: which model → ai-coding-model-guidance / claude-api; retrieval → ai-rag-engineering; eval at scale → llm-evaluation-engineering; adversarial testing → ai-red-teaming; the app → claude-app-engineering / backend-engineering. Requires ravenclaude-core@>=0.7.0.

4 skills · 3 agents

Rdzen weryfikacyjny instaluj-zawsze (11 skilli): routing zadania, ocena wejscia, grounding cytatu, ekstrakcja cytatow, sylogizm/subsumpcja, czerwony zespol, jednoprzebiegowy atak przeciwnika, test zimnego czytelnika (pierwsze wrazenie sedziego), wiernosc deliverable, ocena outputu, paczka audytowa AI Act art. 12. Neutralny jurysdykcyjnie, bez konektorow, bez wlasnych polaczen wychodzacych.

11 skills

Claude reasons better with structure. This skill adds Chain-of-Thought to every question: automatic decomposition, verifiable steps, anti-hallucination guards. Three modes — Flash, Normal, Deep — scaled to the complexity of your question.

1 command

用 COT 非商業部位變化，量化對沖基金在農產品期貨的資金流向，並把出口需求、USDA 數據、美元/原油/金屬等宏觀風向整合成可交易的敘事與訊號。

1 skill