Single-agent mission orchestrator. Decomposes a mission into tasks, spawns one worker per task using the default model, verifies exit criteria, and chains tasks with retry. No critic pattern — each worker self-verifies. Simple, fast, works with any model config.
This skill teaches the core development patterns and conventions used in the agentshield TypeScript codebase. It covers file naming, import/export styles, commit message conventions, and testing patterns. This guide is designed to help contributors quickly align with the project's established practices and workflows.
Manage releases for this project. Validates changelog, installs git hooks, and cuts releases. Use when user says "/release", "release 1.0.5", "cut a release", or asks about the release process. NOT auto-invoked by the model.
Discover, install, and configure Vercel Marketplace integrations (databases, auth, logging, etc.). Use when adding third-party services to a Vercel project.
Generate and iterate on UIs/apps with Vercel v0 via the v0 Platform API. Use when building a frontend with v0, calling the v0 API/SDK, or spending v0 credits.
Conventions and safe-change rules for the PotBot Anchor programs (potvault, potduel). Use when adding or modifying instructions, accounts, PDAs, or governance logic in packages/program, when touching account layouts/migrations, or wiring program changes into the SDK and the web mock store. Triggers on "add…
Deploy and verify the PotBot potvault Anchor program from devnet to Solana mainnet. Use when the user says "deploy", "go to mainnet", "задеплой", "запусти в мейннет", "ship the program", needs to fund the executor wallet, verify a program ID, or run the end-to-end mainnet smoke test. The mainnet deploy is the funded…
Expert developer context for the PotBot v2 project — group trading vaults on Solana (Solana Frontier 2026). ALWAYS use when the user mentions potbot, pot vault, anchor program, devnet/mainnet deploy, the hackathon, strategy vaults, governance, leaderboard, AI agent, personal voters, MCP server, money tree /…
Local video clipping automation for /clippos requests, attached video files, video links, social clip generation, captioned shorts, crop/framing, and rendered 9:16, 1:1, or 16:9 exports. Use this skill whenever the user wants an agent to turn a video into high-potential clips, score clips with the current harness…
This skill should be used when the user needs to interact with the ScuttlePay API to manage an AI agent's USDC wallet. Typical triggers: "check my ScuttlePay balance", "search products on ScuttlePay", "buy a product with USDC", "view my transaction history", "get my wallet address", "list merchants", or any request to…
Use when starting or joining a team project or competition repository and collaboration rules need to be set up — commit/PR/branch conventions, agent context files, GitHub merge settings. Triggers — new hackathon/competition repo, "collaboration rules", "commit convention", "PR rules", "set up team repo", team…
Use when starting or joining a team project or competition repository and collaboration rules need to be set up — commit/PR/branch conventions, agent context files, GitHub merge settings. Triggers — new hackathon/공모전/대회 repo, "협업 규칙", "커밋 규칙 세팅", "PR 규칙", "팀 레포 준비", team develops with coding agents (Claude Code…
INTERNAL ONLY. Invoked by /jelou-build-workflow with a plan fragment. Do NOT trigger this skill directly — it requires a parent-issued envelope and will refuse without one. For standalone node creation, use /jelou-build-workflow.
Build, test, and deploy serverless TypeScript functions on Jelou Functions (the define/app API + jelou CLI). Use when writing or modifying function code, configuring cron or MCP tools, multi-tool functions, runtime-token auth, session memory, scheduled/deferred invocations, webhook verification, the template registry…
Install, authenticate, update-check, and navigate the Jelou CLI (jelou). Use when installing jelou, checking/upgrading the Jelou CLI, running jelou login, managing profiles (auth list/switch/remove), troubleshooting 'jelou: command not found', switching companies, using --json output for scripts and agents…
Run one evidence-driven design research iteration — measure real rendered reference pages with Playwright, aggregate them into numeric craft bands, score our own generated output against those bands, fix the widest gap, and record the loop. Use this before changing anything in the design engine, and once per session…
Design or redesign premium websites/webapps (SaaS, marketing, studio, fintech, consumer) customized to product features with restrained motion. Auto-triggers website-domain-research first (LoadPrior → gap → walkthrough → IA → training), then analyzes requirements, routes a skill graph, applies aesthetic lean profiles…
Auto-trigger mandatory research gate before building any sport website (cricket, football, hockey, tennis, live scores, matchday, league site). Captures cultural vernacular, fan access modes, format lenses, category gaps, and score-spine UX — then seeds the design engine via sport-vernacular packs. Extends…
How to compute what an access grant actually unlocks, rather than what its policy document says. Load before writing or reviewing any ACCESSGRANT change.
Outbound messages to many real humans — suppression, consent, deduplication by person, sender reputation and quiet hours. Load before writing or reviewing any COMMSBLAST change.
Refunds, payouts and adjustments — idempotency, double-payment detection, disputes, and why money never gets an undo certificate. Load before writing or reviewing any MONEYMOVEMENT change.
Use this skill for data incidents and for DataHub questions large enough that loading the metadata would flood the context window. It does two things: myceliumtriage produces a complete on-call report for an alert — cause, severity with its inputs, blast radius, owner, next actions, and what could not be determined…