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 Kvxw1105/agent-gateway-mcp --skill agent-gateway-orchestratorgit clone --depth 1 https://github.com/Kvxw1105/agent-gateway-mcpWrote 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/kvxw1105/agent-gateway-mcp/agent-gateway-orchestrator)<a href="https://agentmods.dev/skills/kvxw1105/agent-gateway-mcp/agent-gateway-orchestrator"><img src="https://agentmods.dev/badge/skills/kvxw1105/agent-gateway-mcp/agent-gateway-orchestrator/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/kvxw1105/agent-gateway-mcp/agent-gateway-orchestrator"><img src="https://agentmods.dev/badge/skills/kvxw1105/agent-gateway-mcp/agent-gateway-orchestrator.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 | $0.00061 | $0.00604 |
| Opus 5 | $0.00030 | $0.00302 |
| Sonnet 5 | $0.00012 | $0.00121 |
| Haiku 4.5 | $0.00006 | $0.00060 |
Grade A, and why
agent-gateway-orchestrator 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 11d 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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Gateway Orchestrator
Act as the controller and use the agent-gateway MCP for process lifecycle. Do not recreate process management in prompts or inject into an open interactive TUI.
Start
- Call
agents_listfor current providers and capabilities. - Inspect the target directory and Git state.
- Read
references/routing.mdwhen selecting agents. Readreferences/setup.mdwhen installing or maintaining the gateway. - Skip unavailable or unauthenticated providers; never disguise an incompatible endpoint as another provider's API.
Select a mode
economy: default. Start with one lower-cost agent and escalate only after failure or failed validation.compare: use only when the user requests a harness comparison. Keep the prompt and output short; use identical inputs and permissions.critical: let one agent implement, a different agent review read-only, and the controller perform final validation.
If the user says quota or budget is low, force economy unless they explicitly request a bounded comparison.
Execute
- Define a testable result, allowed files, permission, output limit, timeout, and prohibited actions.
- Call
agents_spawn; retain task ID, provider, session ID, and working directory. - Poll with bounded
agents_waitcalls and read incrementalagents_logs. - Use
agents_cancelon timeout, drift, or abnormal cost. Do not retry the same failing route indefinitely. - Use
agents_resumeonly to start a new controlled non-interactive process. - Validate locally. An agent's success claim is not completion evidence.
Agents report through Gateway logs and results to the controller. They do not communicate directly by default. For staged collaboration, pass only the previous agent's concise result to the next agent.
Safety
- Default to
read-only. - Protect existing uncommitted work.
- Parallel writers require separate Git worktrees and
isolated_worktree=true. - Never allow multiple agents to write the same ordinary directory.
- Never put credentials in prompts, repositories, MCP config, or logs.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 11d ago First seen · 58 lines · 61 tokens per session scan A ba6725d325f9
agent-gateway-orchestrator is a skill published in the GitHub repository Kvxw1105/agent-gateway-mcp (0 stars, last pushed 21d ago), licensed Apache-2.0. It adds 61 tokens to every session and 604 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
skill-router
Use first when a task may benefit from specialized capability in a large local skill library. Search indexed metadata, select an explainable ranked candidate, then fetch only the exact SHA-256 revision returned by search.
opencli-sitemap-author
Use when creating or maintaining OpenCLI site sitemaps: agent-facing navigation, page-state, action, workflow, API-reference, pitfall, and fallback knowledge for a website. Use after browser exploration discovers durable site context, when a sitemap is stale, or when promoting local site knowledge into the repo.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
interview
Ask one useful structured question at a time only when material product/implementation choices are genuinely missing; remember answers and produce a brief/spec. Discoverable facts should be investigated instead of asked.
recipe-create-meet-space
Create a Google Meet meeting space and share the join link.
atmos-validation
Validate Atmos projects, components, arbitrary JSON Schema inputs, EditorConfig, and GitHub Actions; use affected-file selection and native CI annotations.