mows-core

Plugin mows-core from marketplace mows-harness in alhe99/mows-harness. Commands read from the repo's manifest — 7 days ago.

Skills (13)

• build-with-agent-team

  Build a project using Claude Code Agent Teams with tmux split panes. Takes a plan document path and optional team size. Use when you want multiple agents collaborating on a build.

• context7-mcp

  This skill should be used when the user asks about libraries, frameworks, API references, or needs code examples. Activates for setup questions, code generation involving libraries, or mentions of specific frameworks like React, Vue, Next.js, Prisma, Supabase, etc.

• discord

  Send files or messages from a session to the user's Discord server. Triggers on "send to discord", "ship to discord", or "/discord".

• graphify

  any input (code, docs, papers, images) - knowledge graph - clustered communities - HTML + JSON + audit report

• handoff

  Write a context-handoff brief (HANDOFF.md) into a target project directory so a NEW Claude Code session started there resumes with full context of the current work. Use when the user wants to move/switch/jump to another project or repo and continue where this session left off, or says things like 'm

• oci-free-check

  Audit the Oracle Cloud tenancy to confirm everything is still Always Free and $0.00 — checks actual spend, A1 OCPU/RAM limits, the 200 GB storage cap, volume VPU tier, backups, every billable service type, the budget tripwire, and idle-reclamation risk. Use when the user asks whether their OCI VM is

• prompt-master

  Generates optimized prompts for any AI tool. Use when writing, fixing, improving, or adapting a prompt for LLM, Cursor, Midjourney, image AI, video AI, coding agents, or any other AI tool.

• qa

  Run on-demand agent-driven browser QA journeys defined as markdown in docs/qa/journeys/. Headless by default; supports a noVNC manual handoff for login/OAuth/takeover steps. Use when the user says /qa, "qa run <name>", "qa init", "qa list", "run the QA journey", or asks to QA a web flow end-to-end o

• ship-spec

  Use this skill whenever the user invokes `/ship-spec`, says "ship this feature", "ship the spec", "run the full pipeline", "implement and deploy", or otherwise asks to take a spec/branch from code to production autonomously through implement + QA + deploy phases. Also trigger when the user describes

• spec-driven-orchestration

  Use this skill whenever the user wants to set up spec-driven or plan-driven agent orchestration in a Claude Code project — building a workflow where a spec.md / plan.md / tasks.md / SPEC.md file gets fanned out to specialist sub-agents. Trigger on phrases like "set up an orchestrator", "build a spec

• spec-detector

  Use whenever the user asks to implement a spec, run the plan, build the tasks, work through tasks.md, or mentions spec.md / plan.md / tasks.md / SPEC.md. Detects which spec artifacts exist in the repo and routes the conversation to @spec-orchestrator. Make sure to use this skill whenever spec/plan/t

• stack-health

  Full stack-integrity health check for a self-hosted Claude Code harness box (reference deployment: Oracle Cloud A1, aarch64) — verifies every systemd service, any configured extra service (HARNESS_EXTRA_SERVICES, e.g. a Discord bot gateway) and its Node engine compliance, listening ports, local endp

• sync-to-vps

  Full-copy a local project from this machine to a remote VPS at the same home-relative path — project files, Claude Code history (transcripts with cwd rewritten for the VPS), session todos, plus an additive push of ~/.claude skills/commands/agents. Use when the user asks to sync/copy/move a project t

Slash commands (9)

• /mows-core:agent

  Generate a professional pull request description from this session's code changes (analyzes file modifications, tests, and docs; outputs copy-paste-ready markdown — no arguments)

• /mows-core:efforts

  Pick/set your default effort level from mobile (rewrites settings.json — the native /effort doesn't work over remote control)

• /mows-core:implement-spec

  Implement a spec file. Reads spec.md (or the named file), breaks it into tasks, and executes them in order with verification checkpoints.

• /mows-core:models

  Pick/set your default model from mobile (rewrites settings.json — the native /model doesn't work over remote control)

• /mows-core:open

  Start a FRESH session in any project directory from your phone — spawns a per-project remote-control listener (the fix for new sessions being stuck at the base path)

• /mows-core:progress

  Mobile-friendly background-tasks view — reports running workflows & background jobs as text (the web "Background tasks" panel can't render on mobile)

• /mows-core:ship-spec

  Take a spec file (or feature branch) from code to production through the full pipeline — implement, QA (gates + visual + regression), open PR train, watch CI, merge, release PR, tag, deploy, monitor. Uses a four-role agent team (Lead, Builder, QA, Deploy).

• /mows-core:switch

  Change model/effort for THIS session — same conversation, applies NOW (re-hosts the session; the fix for /model & /effort not working over remote control)

• /mows-core:work

  Work in a project directory over remote control — the absolute-path substitute for /cd (which doesn't work on mobile)

Subagents (1)

• pr-summary-generator

  Use this agent when you need to create a professional pull request description after completing code changes. Examples: <example>Context: User has finished implementing a new feature and wants to create a PR description. user: 'I've finished adding the user authentication feature. Can you help me cr