Productivity

Wire a repository through the Caveman Cloud gateway so every LLM request is measured, with no behavior change. Use for "set up caveman" or adding LLM spend observability.

Turn a Caveman optimization observation into an operator-chosen candidate with a paired baseline evaluation. Use when asked to inspect or evaluate a Caveman optimization report. Needs explicit approval.

Stop. That last message did not land: re-pitch it.

Audits and repairs an Obsidian vault's structural health — broken [[wikilinks]], ambiguous duplicate filenames, missing or malformed frontmatter, stale status markers, and hub notes whose state sections have gone cold. Use when the user asks about vault health, broken links, vault cleanup, dead links, orphan notes, tag consistency, or says the vault feels messy or out of date. Trigger phrases include "vault health", "broken links", "clean up the vault", "audit the vault", "is the vault current". Read-only by default; every repair is proposed before it is made.

Triage an email inbox — sort what needs a reply, draft the replies, flag what needs a decision, and surface what can be ignored or unsubscribed. Use when the user says triage my inbox, do my email, what's in my inbox, or clean up my email.

Authors a workstream brief for dispatching multiple parallel build agents across one codebase, with per-lane file ownership, shared-ground warnings, and a derived DO-NOT list. Use whenever work is about to be split across concurrent agents, workers, lanes, or waves — even if the user does not say "write a brief". Trigger phrases include "fan out", "parallel agents", "orchestrate this", "split into lanes", "workstreams", "run these in parallel", "review-gated build". Produces the brief; does not dispatch.

Explicit-invoke divergent-ideation booster (a 5-frame fanout run natively via the Agent tool). Use ONLY when the user types /fanout or says "run a fanout" / "fanout this" / "give me frames on X". Runs 5 parallel cognitive-frame ideation agents + a critic pass on an OPEN-ENDED problem with many valid answers (design, API surface, naming, strategy, hypothesis generation, "give me a few ways to..."). Do NOT use for CONVERGENT work -- root-cause debugging, "why did X break", verifying a fix, anything with one checkable answer -- single-shot is better and cheaper there. Costs ~7x the tokens of a single-shot answer for a small breadth/trap-coverage gain, so it is gated, never automatic.

Run a structured multi-perspective review of a decision instead of getting one agreeable answer. Use when the user says run a council, get me perspectives on this, should I do X, stress-test this decision, or is this a good idea.

Write a durable record of a work session — what was built, root-caused, or shipped — into the project's notes. Use when the user says log this session, write a session log, record this, or when a substantial piece of work has just finished and needs to be handed off to a future session.

Turn an offer into a concrete multi-step sales funnel spec - page-by-page structure, price ladder, copy outline and the metrics each step must hit. Use when asked to build a sales funnel, marketing funnel, landing page flow, lead magnet funnel, webinar funnel, tripwire or VSL funnel, when planning a product launch page flow, or when someone asks "what pages do I need" for selling something online.

Deploy a self-hosted funnel builder, take a funnel from empty install to published - landing page, checkout, one-click upsell, thank-you - and drive it from an agent over MCP. Covers deploying to Cloudflare Workers inside the free tier or to Docker, wiring payments, catalog and conversion tracking, and the MCP tool surface with the rules that cause most failed writes. Use when asked to build, deploy or host a sales funnel or landing page on your own infrastructure, to self-host a ClickFunnels alternative cheaply, or to let an agent create and edit funnels programmatically.

Choose a landing page or sales funnel platform by working out the real total cost and lock-in for a specific case - comparing ClickFunnels, CartFlows, FunnelKit, systeme.io, GoHighLevel, Shopify apps, hand-rolled pages and self-hosted open-source options. Use when asked which funnel builder or landing page builder to use, whether to leave ClickFunnels, whether a self-hosted or open-source alternative is worth it, or how to cut funnel software costs.

Generate valid fancy-kanban boards — markdown-embedded kanban boards with typed fields, swimlanes, workflow rules, and card face configuration for Obsidian and compatible tools

Use when the user wants to post or schedule to a Telegram channel or group through Publora, using a bot. Covers Telegram markdown, silent delivery, link previews and forward protection. Bot API caps media captions at 1,024 characters and videos at 50 MB. Not for other platforms (each has its own skill).

当某件事靠模型自己完成预计会消耗大量 token，但人类只需几秒、不费力气就能解决时，用提问把它交给人类，节省 token 与时间。适用于需要反复猜测、多轮试错、大范围搜索，或只有用户掌握答案的时刻。Use when a subtask would burn many tokens yet a human can settle it effortlessly.

Run any question, idea, or decision through a council of 5 AI advisors who independently analyze it, peer-review each other anonymously, and synthesize a final verdict. Based on Karpathy's LLM Council methodology. MANDATORY TRIGGERS: 'council this', 'run the council', 'war room this', 'pressure-test this', 'stress-test this', 'debate this'. STRONG TRIGGERS (use when combined with a real decision or tradeoff): 'should I X or Y', 'which option', 'what would you do', 'is this the right move', 'validate this', 'get multiple perspectives', 'I can't decide', 'I'm torn between'. Do NOT trigger on simple yes/no questions, factual lookups, or casual 'should I' without a meaningful tradeoff (e.g. 'should I use markdown' is not a council question). DO trigger when the user presents a genuine decision with stakes, multiple options, and context that suggests they want it pressure-tested from multiple angles.

Render Markdown from templates and structured data using Knap CLI. Use when the user asks to apply a Knap template, turn JSON or CSV data into notes, batch-generate Markdown files, or format Defuddle output into a note.

Use this skill after completing a coding, deployment, config, database, or release task, when the user says 记一下 / 记一条 / 沉淀一下 / 总结到知识库 / 更新知识库 / 记到 Obsidian / 写发版文档, or asks where this week's finished work should be recorded. Records only landed, release-relevant facts (scripts, configs, paths, validation, rollback) as durable notes in the Obsidian vault and maintains the weekly release note (发版) — only when there are landed facts, not plans or speculation. Do NOT use for note editing, creating, moving, renaming, reviewing, MOC maintenance, weekly reports, or general what-I-did-this-week summaries — use `obsidian-personal-knowledge-base` for those. Do NOT use for brainstorming or speculation.

Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress tangents, give specific time estimates, make wins visible. Invoke with /i-have-adhd; stays on until "stop adhd mode".

Four output-discipline rules for AI agents: Implicit-Need Fill (surface the request behind the request before answering), Structured Output (conclusion first, no rambling), 3-Option-Plus-Premium (non-code tasks get >=3 compared options plus one premium recommendation), Assumption-Circuit-Breaker (pre-send logic check, explicit assumptions, coverage statement, fallback plan). Invoke with /four-rules; L3 blocks are internal-only, never upload.

Claude skills by Alex Zapadenko — free SKILL.md files, mirrored from product.inc/skills

Reusable Codex skill for CPWRR project review and interview storytelling

Hook-free Codex Skill for proportional development workflows and cross-session continuity

Codex skill for cautious speaker-labeled meeting transcripts and grounded summaries.