Loading...
Loading...
Found 144 Skills
Search and maintain ~/tk, a personal knowledge base of hard-won understanding reused across sessions, such as tool behavior and pitfalls, how repos and services in a workspace connect, and environment facts. Use before working with a tool, repo, or service, when hitting an unexpected error, or when cross-service context is needed. Also use right after figuring out something non-obvious, after a wrong assumption gets corrected, when the user explains how an internal system works, and before the final answer to vote and record.
Route explicit LLM Wiki capture, ingest, integration, or bootstrap work.
Captures and maintains learnings, errors, and corrections. Use when a command fails, the user corrects an assumption, a capability is missing, knowledge is outdated, or a better approach is found. Review relevant saved guidance before major tasks, reuse, and promotion; revalidate it according to what could change and how quickly.
Document a solved problem as one searchable file in the project's docs/solutions/<category>/ so future sessions (and insistir's learnings researcher) find it instead of re-solving it. Three parallel read-only subagents extract context, root cause and fix, and prevention; you write the single file. Use right after a non-trivial fix is confirmed: "that worked", "it's fixed", "problem solved", "document this fix", "write up the solution", "compound this", "ya funcionó", "quedó arreglado", "documenta la solución". NOT for trivial fixes (typos, obvious one-liners), problems still unresolved (keep debugging first), or general docs, READMEs and changelogs.
Constrói uma base de conhecimento (KB) que a IA consegue consultar de verdade: organiza o que você sabe sobre um domínio em conceitos (para entender) e receitas (para fazer), com porta de entrada, consulta rápida e limites de tamanho que mantêm o material utilizável. Cria o índice do projeto na primeira vez e registra cada domínio novo. Recomendada antes de iniciar um fluxo SDD, para as fases seguintes partirem do terreno já mapeado. A revisão do conteúdo é sempre do profissional — a IA organiza e redige, mas quem responde pelo que fica registrado é quem tem a responsabilidade técnica. Use quando o usuário quiser ensinar seu contexto à IA, catalogar leis, normas, manuais, instruções ou políticas internas, documentar um processo, ou reclamar que precisa reexplicar as mesmas coisas em toda conversa.
Manage Yuque (语雀) knowledge bases, documents, and team collaboration through API integration. Supports personal search, weekly reports, knowledge base management, document CRUD, and group collaboration workflows. Based on yuque/yuque-skills.
Manage Obsidian vaults with full support for Obsidian Flavored Markdown — wikilinks, embeds, callouts, YAML properties, Dataview queries, Canvas, and Bases. Organize notes using MOCs (Maps of Content), atomic note principles, and consistent folder/tag taxonomies. Always ask the user where to save before creating notes.
Search the user's wiki for stored knowledge. Use when the user asks "what do I know about X", "find Y in the wiki", before doing external research the wiki may already cover, or when stored context could help the current task and the user hasn't asked — offer once.
Auto-capture technical realizations to Logseq when the user shares how tools or systems actually work. PROACTIVELY invoke without waiting to be asked. Pattern: declarative statements (not questions) revealing expectation vs reality — 'didn't know X does Y', 'turns out Z', '[tool] actually [behavior]', 'figured out why', 'the reason is', 'good to know', 'oh interesting', 'TIL', 'wait so', 'huh'. Skip questions, task requests, trivial fixes, and secondhand info. Manual trigger: user says 'capture/save/document this learning'.
Configuration of the three-layer structure for the Codex second brain. Load when the user says “build the second brain”.
Render a self-contained HTML viewer for a pro-workflow wiki. Pages, sources, claims, seed queue, page-link graph and full-text search all in one file. No external dependencies, no JS framework, S3-uploadable. Use when the user wants to browse a wiki visually, share its current state with someone, audit research progress, or hand off a knowledge base. Inspired by Thariq Shihipar's "Unreasonable Effectiveness of HTML" — favors information density and shareability over markdown-only outputs.
Periodic wiki health check — orphans, missing links, contradictions, stale (wiki→wiki VÀ code→wiki drift), index gaps. Bước 0 là cổng no-op tất định (wiki-sync, 0 token) — code không đổi kể từ neo thì kết luận "wiki current" và dừng sớm hợp lệ.