Standards-compliant code for the Research Project Template — infrastructure or project src/, thin orchestrators, type hints, no mocks. USE WHEN implementing an algorithm, utility, analysis method, optimizer, new module in projects/*/src or infrastructure/, or user says add code following template architecture — even without docs/prompts. Not for end-to-end feature spanning manuscript + pipeline (use template-feature-addition).

Research-to-publication orchestrator for template projects: research, write, verify, review, revise, reproduce, validate, and finalize. USE WHEN the user wants the whole paper workflow or enters midstream with an existing paper or reviewer comments.

Template-native manuscript planning, outline, drafting, revision, formatting, citation check, and AI-use disclosure routing. USE WHEN the user asks to write, outline, revise, format, or prepare a paper inside the Research Project Template.

Read-only manuscript review, methodology review, re-review, and calibration workflow for template projects. USE WHEN the user asks for peer review, referee-style critique, methods/statistics review, or verification that revisions addressed prior comments.

Router and workflow hub for the Research Project Template — code, tests, manuscripts, pipeline, validation, docs, and reproducibility. USE WHEN the user wants to audit the repo, debug ./run.sh or execute_pipeline.py, add a feature, write tests, scaffold a manuscript, verify claims, fix cross-references, run validation gates, refactor, or prepare a Zenodo release, research a topic, write/review/revise a paper, or run a research-to-publication workflow — even if they never say "prompt", "skill", or docs/prompts. Also USE WHEN intent is ambiguous ("something wrong with my manuscript", "full health check", "fix the pipeline") — pick ONE child skill from the routing table and follow it. Do NOT use for generic homework, unrelated app security scans, or casual PDF summarization.

Open Gauss in this workspace is the math-inc/OpenGauss gauss CLI for Lean session tooling—not Huawei OpenGauss DBMS. See docs/reference/opengauss-naming.md; for the public Lean exemplar see projects/template_active_inference/lean/.

Orchestrate continual learning by delegating transcript mining to agents-memory-updater and writing durable memory to local JSON (never root AGENTS.md Learned sections). USE WHEN continual learning, mine prior chats, maintain agent memory, or the stop hook triggers this skill.

Summarize or extract text/transcripts from URLs, podcasts, and local files (great fallback for “transcribe this YouTube/video”).

Create or update AgentSkills. Use when designing, structuring, or packaging skills with scripts, references, and assets.

Two-layer memory system with SQLite-based recall.

Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.

Schedule reminders and recurring tasks.

Comprehensive OpenTUI skill for building terminal user interfaces. Covers the core imperative API, React reconciler, and Solid reconciler. Use for any TUI development task including components, layout, keyboard handling, animations, and testing.

Answer questions about the AI SDK and help build AI-powered features. Use when developers: (1) Ask about AI SDK functions like generateText, streamText, ToolLoopAgent, embed, or tools, (2) Want to build AI agents, chatbots, RAG systems, or text generation features, (3) Have questions about AI providers (OpenAI, Anthropic, Google, etc.), streaming, tool calling, structured output, or embeddings, (4) Use React hooks like useChat or useCompletion. Triggers on: "AI SDK", "Vercel AI SDK", "generateText", "streamText", "add AI to my app", "build an agent", "tool calling", "structured output", "useChat".

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

Activate when Cowork is opened and watchlist monitoring cadence has elapsed. Activate when user says "check my watchlist", "what's new on my watchlist", "run watchlist", "any updates", "watchlist digest", or configures monitoring via /curiosity-stack:watchlist command. ALSO activate immediately — opening the watchlist UI — when user says any of: "add to watchlist", "add this to watchlist", "watch this", "monitor this", "put this on my watchlist", "track this", "add [topic] to watchlist". When triggered by an add request, open the UI with the topic pre-filled. Do NOT activate on every session — check last_run date in local.md first.

Activate when user runs /curiosity-stack:stress-test or says "stress test this", "challenge this thesis", "what could go wrong", "bull vs bear", or after a decomposition output when user wants to pressure-test their conclusion.

Activate when user asks where to find information at any layer, or when they are stuck and need pointers to go deeper. Always check connected personal sources first.

Activate at Layer 1 and Layer 5 of any decomposition to surface the user's highest-rated sources for the current sector and layer. Activate after any session to prompt source rating. Activate when user says "rate this source", "how good is [source]", "which sources should I use for [sector]", or "update my sources". Never activate mid-conversation unless at L1 or L5 transition.

Activate on first use or when user says "setup", "configure", "connect my notes", "link my drive", "how do I connect", or on first session with no prior configuration. Also reads curiosity-stack.local.md if it exists — skip steps already answered there.

ALWAYS active. Apply to every single response, every output, every command in this plugin. This skill takes absolute priority over all other skills. Never disabled, never overridden. Intercept and redirect any request that crosses into investment advice territory.

Activate when user says "I don't know what to research", "give me ideas", "what can I decompose", "show me scenarios", "browse topics", "not sure where to start", "what topics are available", or runs /curiosity-stack:scenarios. Also activate on first session if local.md session_count is 0 or empty — surface the library automatically after the welcome message. Never activate mid-decomposition.

Activate after every completed decomposition to generate 5 curated reading recommendations specific to the topic and layers explored. Also activate when user asks "what should I read", "where do I go deeper", "give me links", "reading list", or "resources on this".

Activate when a user selects their output format after completing a decomposition. Generates exactly one of: Value Chain diagram, interactive Mindmap, Research Note, or Research Brief PDF. Never generate all — only what was selected. Always include SEBI disclaimer, attribution badge, and invisible source comment.