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 enwong93-sketch/devspace-ultra --skill devspace-ultra-setupgit clone --depth 1 https://github.com/enwong93-sketch/devspace-ultraWrote 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/enwong93-sketch/devspace-ultra/devspace-ultra-setup)<a href="https://agentmods.dev/skills/enwong93-sketch/devspace-ultra/devspace-ultra-setup"><img src="https://agentmods.dev/badge/skills/enwong93-sketch/devspace-ultra/devspace-ultra-setup/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/enwong93-sketch/devspace-ultra/devspace-ultra-setup"><img src="https://agentmods.dev/badge/skills/enwong93-sketch/devspace-ultra/devspace-ultra-setup.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 | $0.00086 | $0.03046 |
| Opus 5.5 | $0.00034 | $0.01218 |
| Sonnet 5.5 | $0.00017 | $0.00609 |
| Haiku 4.5 | $0.00009 | $0.00305 |
Grade A, and why
devspace-ultra-setup 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 1mo 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 — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.
DevSpace Ultra guided setup
Use this Skill as an interactive installation and recovery runbook. The Agent executes every safe local command it can execute, verifies each result itself, and asks the user only for account, router, or physical-network actions that cannot be completed locally. Do not merely paste a long command and leave the user to diagnose it.
Core product contract
- The primary client and product acceptance target is ChatGPT Classic on Windows.
- Codex may act as the installer or repair executor, but Codex is not the product client and a working Codex connection is never DevSpace Ultra acceptance.
- The core outcome is one OAuth-protected ChatGPT Classic MCP connection that exposes verified local development tools.
- One ChatGPT account may contain several DevSpace connections for different computers. Each installation needs its own DDNS/tunnel URL, persisted
serverInstanceId, OAuth resource, and Connector entry. Never reuse or test another computer's DevSpace connection as acceptance for this machine. - Long-context checkpoint, handoff and recovery can keep a project moving beyond one conversation's usable context. This is continuity, not a literal increase or bypass of model, account, rate or token limits; never advertise “unlimited tokens” as a technical entitlement.
- Local setup, DDNS, Caddy, health checks, optional Multi-Main, Goal/Rescue and extra capabilities are incomplete until the ChatGPT Classic MCP connection, OAuth and real read/write/reconnect tests pass.
Mandatory order — never omit or silently reorder
Infrastructure checks are prerequisites, not the product goal. Resume at the earliest incomplete gate. If the public endpoint already passes external checks, go directly to ChatGPT Classic connection and OAuth; do not reopen the router, reinstall Caddy or redo DNS.
- Inspect and preserve existing state.
- Prepare or verify local Gateway/Core and the public OAuth-protected
/mcpendpoint. - In ChatGPT Classic, enable Developer mode, create the exact MCP connection and complete OAuth.
- Confirm the connection URL is this machine's exact public
/mcpresource and its reportedserverInstanceIdmatches this installation. A similarly named connection for another computer is a separate authority. - A new conversation's first
devspace_progress_reportmay briefly returnpending=truewhile the hidden exact-page relay confirms ownership. Do not ask the Agent or user to calldevspace_progress_bind; continue unrelated safe workspace work and verify the narration card before claiming the report succeeded. The legacy bind handler exists only for cached hidden-App compatibility.
- Confirm the connection URL is this machine's exact public
- Scan or refresh the authenticated tool catalogue and verify all required development tools.
- Enable the connection in a fresh ChatGPT Classic conversation.
- Run a real read test.
- Run an explicitly authorized disposable write/edit/command test and verify it on disk.
- Reconnect or open another fresh Classic conversation and prove authenticated tools persist.
- Report every gate as pass/fail. A missing or unverified row means installation is incomplete.
What ships with it
1 file 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.
- 1mo ago Changed 6a5b63f99bdd
- 1mo ago Changed · +1 lines 353874aad9a4
- 1mo ago Changed · +32 lines f55e60fb43e6
- 1mo ago Changed f0b0fd6009d3
- 1mo ago Changed · +15 lines c10cedb63cd9
- 1mo ago Changed 1e02d85a7551
- 1mo ago Changed baeeb37d0c7e
- 1mo ago First seen · 134 lines · 86 tokens per session scan A 26df49283024
devspace-ultra-setup is a skill published in the GitHub repository enwong93-sketch/devspace-ultra (58 stars, last pushed 1mo ago), licensed MIT. It adds 86 tokens to every session and 3,046 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-10.
devspace-ultra-setup compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| devspace-ultra-setupenwong93-sketch/devspace-ultra | Skill | 58 | 1mo ago | MIT | A | 86 | Codex |
| gke-compute-classesgoogle/skills | Skill | 22k | yesterday | Apache-2.0 | A | 83 | Claude Code, Codex |
| gke-reliabilitygoogle/skills | Skill | 22k | yesterday | Apache-2.0 | A | 85 | Claude Code, Codex |
| gke-workload-securitygoogle/skills | Skill | 22k | yesterday | Apache-2.0 | A | 171 | Claude Code, Codex |
| cloud-architectJeffallan/claude-skills | Skill | 12k | 7d ago | MIT | A | 71 | Claude Code |
| nemo-automodel-launcher-configNVIDIA/skills | Skill | 3.9k | yesterday | Apache-2.0 | A | 30 | Claude Code |
| azure-mgmt-botservice-dotnetmicrosoft/skills | Skill | 3.2k | yesterday | MIT | A | 78 | Claude Code |