Skill Claude Code
Watch, babysit, audit, review, compare, or fix another agent's work from a session, transcript, PR, branch, log, or pasted run summary. Reconstruct the ask, inspect changes and verification, report gaps, optionally make scoped fixes when authorized.
Skill Claude Code
Business analysis: requirements elicitation, process models, user stories, gap analysis, BRD sections, acceptance criteria, stakeholder maps. Deepiri product context.
Skill Claude Code
Define operating values and hiring bar — culture as constraints on decisions, Deepiri team context.
Skill Claude Code
Competitor matrix, pricing, positioning gaps — uses research CSVs and web search.
Skill Claude Code
React composition patterns that scale. Use when refactoring boolean-prop proliferation, building flexible component libraries, or designing reusable APIs — compound components, render props, context providers, React 19 APIs.
Skill Claude Code
Get an independent second opinion on risky code, architecture, or a plan from a different model/provider than the one that produced it, so a shared blind spot in one model's training doesn't sail through review unchallenged. Use before merging security-sensitive changes, irreversible migrations, or any diff the author…
Skill Claude Code
Validate datasets and training data — schema, nulls, duplicates, source freshness, PII.
Skill Claude Code
Elite demo idea portfolio generator — 25–100 novel, demonstrable concepts with architecture, wow factor scoring, and curated top-10 rankings by audience.
Skill Claude Code
Technical docs for Deepiri — README sections, runbooks, architecture guides.
Skill Claude Code
Use when running Claude Fable on codebase-heavy or token-heavy work and the user wants Fable to orchestrate research, coding, and testing while cheaper subagents do bounded heavy lifting.
Skill Claude Code
Orchestrate high-cost frontier models: keep planning, synthesis, tradeoffs, and final review on the expensive model; delegate research, coding, and testing to cheaper subagents for token-heavy work.
Skill Claude Code
Find where your customers already gather — forums, subs, niches — before building. Pairs with validate-idea and research painpoint CSVs.
Skill Claude Code
Discover and install agent skills when the user asks "how do I do X", "find a skill for X", "is there a skill that can…", or wants to extend capabilities beyond the Axiom catalog.
Skill Claude Code
Run grounded, push-back-capable ideation across one or more parallel agent sessions under a human's raw directive input — trigger-sourced accept/ reject discipline, one committed pick (never a menu), and a human-brokered relay between parallel tracks that shares outcomes, not in-flight reasoning.
Skill Claude Code
Default-alive growth — when to hire, spend, or add scope. Profitability and customer pull over vanity scale.
Skill Claude Code
Working linear algebra toolkit — solving systems, matrix structure, determinants, vector spaces and the four fundamental subspaces, orthogonality and least squares, eigenvalues and diagonalization. Reach for it when a problem reduces to a matrix, when you need to know which decomposition fits, or when reading what a…
Skill Claude Code
Reframe "impossible" blockers with lateral thinking — change the question, outsider-loop workarounds, then a long-term system-level fix so the blocker never exists again. Use when stuck on a dead end, told something cannot be done, or facing physics/logic/policy blockers.
Skill Claude Code
Content-led marketing after product-market fit — audience building without paid spray before manual sales work.
Skill Claude Code
Gut-check any plan through minimalist entrepreneur principles — community, manual-first, ship small, sell first, default alive.
Skill Claude Code
End-to-end product discovery and MVP launch: stakeholder discovery, market and competitive research, structured brainstorming, product briefs, then minimalist-entrepreneur MVP execution (manual → processized → productized). Use for product brief, brainstorm, research, discovery, requirements gathering, problem…
Skill Claude Code
Compare, cross-review, merge, judge, or choose among competing plans from multiple agents (pasted strategies, plan docs, PR descriptions, transcripts). Produce one recommended execution plan with verification gates.
Skill Claude Code
Autonomous progress without routine clarification stops. Use for "plow ahead", "do not stop", "use your best judgment", "keep going until done", "finish while I am away", or "do not ask unless truly blocked". State assumptions, proceed, validate, stop only for true blockers, end with a decision recap.
Skill Claude Code
Find product-market fit with lateral, genius-grade creativity — reframe the market, exploit the underused asset already sitting in your room, then build a self-reinforcing loop so fit compounds instead of fading. Use when early traction is flat, demand feels thin, competitors own the space, or you aren't sure who…
Skill Claude Code
Set initial price, tiers, and path to profitability — cost vs value based, zero-price effect, iterate over time.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: