저장소 README를 불러오는 중…
수집된 skill 61개 중 40개를 표시합니다.
VELTH project preflight checks, absolute rules, git hygiene, environment constants, and the skills map (which of the other ~20 VELTH skills to load for which kind of task). Load this FIRST, at the start of every session (Claude Code or Codex), not only tasks…
원문 언어: 영어
veOS project preflight checks, absolute rules, environment constants, and hard-won traps for every Claude Code session touching velthdev/veos. Load at the start of ANY task in this repo. Distinct from VELTH-proper's velth-preflight — separate repo, separate…
원문 언어: 영어
Project-agnostic skill router across all of Anshu's repos and general work, not just one codebase. Load at the start of EVERY session, automatically, without waiting to be named or for a project-specific preflight to fire first — this is the top-level entry…
원문 언어: 영어
After the loop is green, emit an explicit copy-paste git add + commit + push block IN CHAT for Anshu to run. Cowork never runs git - Anshu commits. Load at the end of an implementation session once velth-loop verification passes. Uses explicit file paths…
원문 언어: 영어
Context and state engineering for VELTH Cowork sessions - how to keep Cowork sharp over long tasks, recover fast across sessions, parallelise safely, and treat ingested content as untrusted. Context degradation is the
원문 언어: 영어
Automated L5 - verify correctness of generated VELTH documents (all 6 doc types, any vertical) without opening a PDF by hand. Deterministic defect checks plus an adversarial DIFFERENT-FAMILY LLM judge plus a localhost/UI propagation smoke. Load whenever a…
원문 언어: 영어
The per-domain loop for VELTH's Supabase-to-Cloud-SQL repository-layer migration (Path B). Encodes the recipe from docs/REPOSITORY_LAYER_MIGRATION.md + PART2 HARDENED with researched best practices (strangler-fig / branch-by-abstraction, asyncpg+pgbouncer…
원문 언어: 영어
The bounded research/implement/verify loop for every VELTH coding task. Encodes plan-implement-verify with iteration caps, a DIFFERENT-FAMILY external judge (maker != checker), checkpoint-and-restore against snowballing, and external-ground-truth-only…
원문 언어: 영어
Code-level security and correctness review that runs inside velth-loop's verify/simplify step - the rung that catches what doc-verify (output correctness) does not. Security review (secrets, PII/GDPR, RLS, injection), property-based invariants for the P0…
원문 언어: 영어
The planning front-end for every non-trivial VELTH task - turn an idea into an approved, testable plan before any code is written. Explore (read-only, with a mandatory ponytail why-does-this-behave-this-way check for existing logic) -> Plan -> confirm ->…
원문 언어: 영어
VELTH mandatory testing strategy for every backend task. Load before writing any test, before any git commit, before any Cowork implementation prompt. Contains all learnings from May 2026 sprint. Universal — applies to every feature, fix, and refactor, not…
원문 언어: 영어
How to read VELTH/veOS's real incident-log skills and adapt their lessons to a DIFFERENT project, instead of trusting a stale hand-copied summary. Load whenever working on a non-VELTH/veOS project and something looks like a familiar shape — "this should be…
원문 언어: 영어
On a VELTH task, prefer velth-preflight's Architecture Boundaries section instead - this generic skill is for other projects or genuine gaps. Guides stable API and interface design. Use when designing APIs, module boundaries, or any public interface. Use when…
원문 언어: 영어
On a VELTH task, prefer velth-prod-verify instead - this generic skill is for other projects or genuine gaps. Automates CI/CD pipeline setup. Use when setting up or modifying build and deployment pipelines. Use when you need to automate quality gates,…
원문 언어: 영어
On a VELTH task, prefer velth-review instead - this generic skill is for other projects or genuine gaps. Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need…
원문 언어: 영어
On a VELTH task, prefer ponytail and ponytail-review instead - this generic skill is for other projects or genuine gaps. Simplifies code for clarity. Use when refactoring code for clarity without changing behavior. Use when code works but is harder to read,…
원문 언어: 영어
On a VELTH task, prefer velth-context instead - this generic skill is for other projects or genuine gaps. Optimizes agent context setup. Use when starting a new session, when agent output quality degrades, when switching between tasks, or when you need to…
원문 언어: 영어
On a VELTH task, prefer velth-gcp-migration instead - this generic skill is for other projects or genuine gaps. Manages deprecation and migration. Use when removing old systems, APIs, or features. Use when migrating users from one implementation to another.…
원문 언어: 영어
On a VELTH task, prefer the existing docs/adr/ convention instead - this generic skill is for other projects or genuine gaps. Records decisions and documentation. Use when you need to document an architecture decision (ADR) or the reasoning behind a design…
원문 언어: 영어
On a VELTH task, prefer impeccable instead - this generic skill is for other projects or genuine gaps. Builds production-quality, accessible, responsive user-facing UIs. Use when building or modifying interfaces and pages, creating components, implementing…
원문 언어: 영어
On a VELTH task, prefer velth-commit-prep instead - this generic skill is for other projects or genuine gaps. Structures git workflow practices. Use when making any code change. Use when committing, branching, resolving conflicts, splitting uncommitted work…
원문 언어: 영어
On a VELTH task, prefer velth-loop instead - this generic skill is for other projects or genuine gaps. Delivers changes incrementally in thin, verifiable slices. Use when implementing any feature or change that touches more than one file, or when picking up…
원문 언어: 영어
On a VELTH task, prefer impeccable for frontend perf instead - this generic skill is for other projects or genuine gaps. Optimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you…
원문 언어: 영어
On a VELTH task, prefer velth-graph-engineering instead - this generic skill is for other projects or genuine gaps. Breaks work into ordered tasks. Use when you have a spec or clear requirements and need to break work into implementable tasks. Use when a task…
원문 언어: 영어
On a VELTH task, prefer velth-review instead - this generic skill is for other projects or genuine gaps. Hardens code against vulnerabilities. Use when auditing an input handler for vulnerabilities, when handling user input, authentication, data storage, or…
원문 언어: 영어
On a VELTH task, prefer velth-prod-verify instead - this generic skill is for other projects or genuine gaps. Prepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need…
원문 언어: 영어
On a VELTH task, prefer research-simulate-encode instead - this generic skill is for other projects or genuine gaps. Grounds every implementation decision in official documentation. Use when you want to verify an approach against the official docs before…
원문 언어: 영어
On a VELTH task, prefer velth-spec instead - this generic skill is for other projects or genuine gaps. Creates specs before coding. Use when starting a new project, feature, or significant change and no specification exists yet. Use when drafting a PRD or…
원문 언어: 영어
On a VELTH task, prefer velth-test-strategy instead - this generic skill is for other projects or genuine gaps. Drives development with tests using the red-green-refactor loop. Use when implementing any logic, fixing any bug, or changing any behavior. Use…
원문 언어: 영어
On a VELTH task, prefer skill-map instead - this generic skill is for other projects or genuine gaps. Discovers and invokes agent skills. Use when starting a session, or when you need to decide which skill or workflow applies to the piece of work at hand.…
원문 언어: 영어
How to point the installed `graphify` skill (a separate, upstream-managed tool at ~/.claude/skills/graphify) usefully at the veOS repo specifically, and which of this project's other skills should consume or feed its output. Load this alongside `graphify`…
원문 언어: 영어
Exploit China's frontier open-source AI, data, research, and engineering ecosystem through evidence-led discovery, code forensics, local reproduction, and global benchmarking before adopting a tested decision.
원문 언어: 영어
Design and validate a research-grounded ML model ladder for time-series, ranking, personalization, or hackathon systems without confusing novelty with evidence.
원문 언어: 영어
Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a…
원문 언어: 영어
Harvest every `ponytail:` comment in the codebase into a debt ledger, so the deliberate shortcuts and deferrals ponytail leaves behind get tracked instead of rotting into "later means never". Use when the user says "ponytail debt", "/ponytail-debt", "what did…
원문 언어: 영어
Code review focused exclusively on over-engineering. Finds what to delete: reinvented standard library, unneeded dependencies, speculative abstractions, dead flexibility. One line per finding: location, what to cut, what replaces it. Use when the user says…
원문 언어: 영어
Use a minimal-change engineering lens without deleting edge-case, safety, security, accessibility, data-integrity, or explicitly requested behavior; route risky architectural choices directly through research-simulate-encode.
원문 언어: 영어
Split a task into judgment-heavy phases (kept on the primary model) and high-volume mechanical phases (delegated to a cheaper/faster tier), to save cost and quota without routing traffic through any third-party proxy. Use when a task is large, has a clear…
원문 언어: 영어
Exploit Silicon Valley and US AI, open-source, research, compute, standards, startup, and commercialization ecosystems through evidence-led discovery, code forensics, local reproduction, and global benchmarking.
원문 언어: 영어
How to actually get and use a real authenticated Playwright session against VELTH production, AND how to do real live UI/UX visual inspection (screenshot + actually look), so a fix touching login-gated behavior (Begehung senden, the guided review, anything…
원문 언어: 영어
이 프롬프트를 복사해 사용 중인 AI 어시스턴트에 입력하세요.
Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/Ansh2508/assistance-system.이 저장소에는 표시할 수 있는 Markdown README가 없습니다. 아래에서 수록된 Skill을 살펴볼 수 있습니다.