aaaAlexanderaaa

Operate the CCHistory CLI — discover sources, sync, query projects/turns/sessions, export/import/backup bundles, and verify restores. Use when the user wants to interact with CCHistory data through the local CLI. Covers all read, preview, and mutating workflows with safety-tiered command guidance.

Find, evaluate, and recommend AI products using the watcha.cn platform API. Use this skill whenever the user asks about AI tools, AI products, AI apps, or wants to discover/compare/evaluate AI products in China or globally. Also use when the user mentions watcha, watcha.cn, or wants product recommendations for specific use cases (e.g., "what's a good AI coding tool?", "find me an AI video generator", "哪个AI写作工具好用"). This skill knows how to search, filter, read reviews, and cross-reference with web sources to give well-rounded product assessments — not just popularity rankings.

Produce expert-quality, opinionated structural analysis of a domain, concept, mechanism, or decision — with explicit stance, explicit boundary, explicit hierarchy, and deliberate omissions. This skill counteracts three default failures of AI on "go deep" tasks: hedging instead of recommending, including everything tangential instead of drawing a boundary, and flattening hierarchy instead of separating core from peripheral. It enforces curate-over-generate (find expert-curated structures first; do not synthesize from training alone), applies a 5-phase process (scope → source curation → alignment → synthesis → self-audit), and adapts output morphology to the task shape: learning roadmap, knowledge map, mechanism analysis, concept audit, or decision guide. Use this skill when the user asks to: • deeply understand / 深入理解 / 深入浅出 a domain or concept • learn from scratch / 从零开始学 / 系统性地学习 / 入门到精通 • get the big picture / 全景图 / 知识体系 / 帮我梳理 X • give me a roadmap / what should I learn about X / end-to-end overvie

Build, debug, and optimize Splunk Enterprise 10.0 SPL searches using a local knowledge base (kb/) and spl_validator. Use for SPL query design, dataflow planning, command/function selection, troubleshooting parse/semantic errors, and iterative validation with AST + flow (warnings are allowed but must be surfaced).