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 agentmods add skills/agentproto/ts/adapter-setup-kitnpx skills add agentproto/ts --skill adapter-setup-kitgit clone --depth 1 https://github.com/agentproto/tsWhat 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 | $0.00191 | $0.01837 |
| Opus 5 | $0.00096 | $0.00919 |
| Sonnet 5 | $0.00038 | $0.00367 |
| Haiku 4.5 | $0.00019 | $0.00184 |
Grade A, and why
adapter-setup-kit 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 2d 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 — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Adapter setup kit (catalog, setup, tunnels)
The upstream of orchestration: before agent_start, you need a ready
adapter; to drive a remote agent or expose a service, you need a tunnel.
This skill covers inventory (list_*), configuration of sensitive providers
(setup_*), and network exposure (tunnel_create). The other orchestration
skills assume all of this is already in place.
The three adapter families
One shared foundation (@agentproto/adapter-kit) catalogs three families, each
with its own list_*:
- CLI agents —
adapter_list(claude-code, hermes, opencode, codex, openclaw…). What you spawn. - Tunnels —
list_tunnel_adapters(cloudflare-quick, cloudflare-named). How you expose. - Browsers —
browser_adapter_list(agents that drive a browser).
All return a uniform 3-state status:
supported= known but not installed (e.g. opencode, codex →version: "not installed"). The package must be installed (@agentproto/adapter-<slug>).available= installed but setup not done (e.g. cloudflare-named:requiresAuth→ needs credentials).ready= installed and configured, usable right away (e.g. claude-code, hermes, cloudflare-quick).
Golden rule: call the list_* BEFORE spawning/opening. Don't guess what
is installed — an agent_start on a supported (not installed) adapter
fails.
Pattern 1 — Catalog (the upstream check)
adapter_list returns for each agent: slug, name, version, protocol
(acp/…), streaming, packageName, models (the list of models the adapter
knows), status, hint. Lived example: claude-code and hermes are
ready; opencode / codex / openclaw are supported
(version: "not installed"). Use models to offer a model choice at spawn,
and status to only offer what is ready.
list_tunnel_adapters additionally returns a capabilities block per
provider: stableUrl, autostart, customDomain, requiresAuth, hasApi —
that's what decides which provider fits the need (stable URL? restart at
boot?).
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.
- 2d ago First seen · 147 lines · 191 tokens per session scan A 9b5e7466cc7c
adapter-setup-kit is a skill published in the GitHub repository agentproto/ts (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 191 tokens to every session and 1,837 once invoked, about $0.0010 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
browser-qa
Validate web applications through the embedded browser with responsive, interaction, console, and screenshot checks.
scene-designer
Activate this skill when users need help creating, modifying, or arranging visual elements in OBS scenes. Triggers include requests like "add a text overlay", "create a new scene", "move my webcam", "resize the game capture", "add a browser source", "arrange my layout", or designing scene compositions. This skill…
ai-agent-browser
业务诊断助手适合市场营销、运营、产品、销售在用户提出“这件事该怎么做”这类问题，需要快速拆解目标、判断重点并形成可执行结果时使用，帮助基于输入材料生成摘要、诊断结论、行动建议和可复用交付物。.
ai-browser-use
业务诊断助手适合市场营销、运营、产品、销售在用户提出“这件事该怎么做”这类问题，需要快速拆解目标、判断重点并形成可执行结果时使用，帮助基于输入材料生成摘要、诊断结论、行动建议和可复用交付物。.
agent-browser
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
docs-writer
Draft or rewrite Strands Agents documentation pages. Use when writing new doc pages, rewriting pages that failed audit, drafting sections for existing pages, or writing blog posts and release notes about Strands. Also triggers on "write a doc", "draft a page", "rewrite the quickstart", "add a tutorial for X"…