A guide to choosing how an agent or person connects to Wingthing, a system for controlling or running agents on local or remote machines. It prioritises local access and adds remote or browser access only when needed.

Use it to select local agent control, a local sandboxed terminal, remote MCP, a self-hosted browser-accessible setup, or an entitled hosted relay.

It prevents choosing a more complicated connection method than the task requires and states which routes are currently supported.

patterns — Skill by ehrlich-b

Install

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.

agentmods

npx agentmods add skills/ehrlich-b/wingthing/patterns

Any agent

npx skills add ehrlich-b/wingthing --skill patterns

Clone the repo

git clone --depth 1 https://github.com/ehrlich-b/wingthing

Made for: Claude Code, Codex.

Per session 27 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 612 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00027	$0.00612
Opus 5	$0.00014	$0.00306
Sonnet 5	$0.00005	$0.00122
Haiku 4.5	$0.00003	$0.00061

Measured 3d ago against content hash ff19617f2b03, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

patterns 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 3d 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.

patterns/SKILL.md · 54 lines

What it actually says

Wingthing patterns

Choose the first supported route that satisfies the request:

Local agent control with stdio MCP: let the current Claude or Codex session manage local sub-agents on existing code and provider logins. Read local sub-agents.
Local sandboxed agent terminal for a person: run and later reattach to an agent on this computer. Read local sandbox.
Direct remote MCP when machines differ: let one AI explicitly select and manage agents on several computers. Read remote orchestration.
Self-hosted roost when a person needs a browser: for one person's remote machine, read personal remote wing. For a team browser host, read shared web roost. To connect an AI to that private roost, read shared roost agents.
Optional entitled hosted relay: only when the account already has hosted-relay access and the selected wing allows it, read hosted browser wing.

Do not present planned features as patterns. In particular, Wingthing does not currently merge independent roosts into one inventory, provide browser-direct free terminals, prepare Git worktrees atomically, or expose remote schedules and delivery targets.

For every pattern, state these six placements before changing the host:

Driver: human or LLM.
Execution: the wing that owns the process.
Workspace: the path and replica available on that wing.
Display: terminal, browser terminal, or preview route.
Credentials: the owner-scoped agent home or delegated credential.
Memory: the durable project record and where it is replicated.

Current Wingthing routes execution. It does not synchronize a workspace or durable memory between wings. A cwd must already exist on the selected wing. Say this plainly when a pattern crosses machines.

Run wt doctor before changing a host. Use wt egg explain <agent> --json to inspect the effective sandbox. Ask MCP clients to call wingthing_capabilities before starting work.

Keep identities separate on shared hosts. The Wingthing login selects the egg owner. Claude, Codex, and other provider logins live in that owner's agent home. Never place provider tokens in MCP arguments, checked-in config, logs, or copied instructions. Preserve existing web-roost users while changing programmatic access unless the user has agreed to a migration.

Files

What ships with it

8 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.

Changes

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.

3d ago First seen · 54 lines · 27 tokens per session scan A ff19617f2b03

Subscribe to this mod's changes

patterns is a skill published in the GitHub repository ehrlich-b/wingthing (52 stars, last pushed 4d ago), licensed MIT. It adds 27 tokens to every session and 612 once invoked, about $0.0001 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.

Building agents

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens