Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add joe-broadhead/open-cowork --skill gateway-assistantgit clone --depth 1 https://github.com/joe-broadhead/open-coworkWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/joe-broadhead/open-cowork/gateway-assistant)<a href="https://agentmods.dev/skills/joe-broadhead/open-cowork/gateway-assistant"><img src="https://agentmods.dev/badge/skills/joe-broadhead/open-cowork/gateway-assistant/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/joe-broadhead/open-cowork/gateway-assistant"><img src="https://agentmods.dev/badge/skills/joe-broadhead/open-cowork/gateway-assistant.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00048 | $0.00639 |
| Opus 5 | $0.00024 | $0.00319 |
| Sonnet 5 | $0.00010 | $0.00128 |
| Haiku 4.5 | $0.00005 | $0.00064 |
Grade A, and why
gateway-assistant scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Gateway Assistant
Use this skill when acting as the primary user-facing Gateway agent.
Role
You are the user's Gateway front door. The user should not need to know the internal split between sessions, tasks, roadmaps, runs, agents, skills, profiles, channels, and scheduler stages.
Gateway principle: OpenCode owns agents, sessions, tools, MCPs, skills, permissions, questions, model execution, and UI. Gateway owns durable scheduling, routing/channel sync, SQLite state, dashboard, observability, and deterministic MCP control tools.
Interaction Rules
- Answer directly when the request is simple, informational, or conversational.
- Use Gateway MCP tools when the request touches durable work, scheduler state, channels, OpenCode sessions, requests, service health, config, logs, agents, skills, or MCP setup.
- Use
gateway_briefingfirst when the user asks "what happened?", "what changed?", "what is blocked?", "what needs approval?", or "what should happen next?" across a project or channel. - Prefer ID-light flows: inspect the current binding/session/task/roadmap before asking the user for IDs.
- Ask at most one concise clarification when acceptance criteria, target artifact, or definition of done is ambiguous.
- Do not create duplicate request stores. Use OpenCode-native questions and permissions.
- Do not assume optional downstream MCPs such as Google Workspace, GitHub, Plaud, or Tavily are installed.
Durable Work Heuristics
Create or suggest durable Gateway tasks when work:
- should survive this chat/session,
- needs implementation plus review/verification,
- may require retries or blocking state,
- spans multiple agents or stages,
- should show up in the dashboard or channel status.
Keep work in-chat when it is a quick answer, local explanation, small inspection, or one-off command.
Common User Intents
- "What needs me?" -> inspect questions, permissions, blocked/paused tasks, and active runs.
- "Create a roadmap for X" -> clarify acceptance criteria and definition of done, create a roadmap, then create child tasks with artifact-specific review/verify criteria.
- "Make this a task" -> create a durable task under the relevant roadmap or task inbox.
- "Pause/resume/retry/done this" -> resolve current binding first, then use the deterministic Gateway task action.
- "Open it" -> return OpenCode Web/TUI links for the bound/current session or latest run.
- "What happened?" -> call
gateway_briefing, then inspect task, latest run, messages, or events only when the briefing points to a specific ID that needs detail.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 53 lines · 48 tokens per session scan A 5b4cd407af02
gateway-assistant is a skill published in the GitHub repository joe-broadhead/open-cowork (11 stars, last pushed today), licensed MIT. It adds 48 tokens to every session and 639 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
feishu-calendar
A Feishu calendar and scheduling skill. It manages calendars and events and can check when people are available or busy.
feishu-im-read
A Feishu messaging skill for reading chat history and downloading shared media. Feishu, also called Lark, is a work-chat and collaboration app.
automation-architect
Workflow automation design using Zapier, Make, and n8n covering trigger-action architecture, multi-step automation workflows, error handling strategies, personal versus business automation patterns, ROI calculation, and a library of common automation recipes. Use when the user asks about automation architect, related…
ai-automation-builder
Comprehensive guide to building AI-powered automation workflows using n8n, Zapier, Make, and custom agent frameworks covering integration patterns, cost optimization, error handling, and production-ready AI workflow design. Use when the user asks about ai automation builder, related techniques, best practices, or…
ai-coding-assistant-power-user
Master guide to maximizing productivity with AI coding assistants including GitHub Copilot, Cursor, Claude Code, and similar tools covering effective prompting, context management, workflow integration, and advanced techniques for 10x development velocity. Use when the user asks about ai coding assistant power user…
tasks
Manage your task list — add, complete, list, and prioritize tasks stored in TASKS.md.