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 oxy-hq/oxygen --skill oxy-customer-apps-perfgit clone --depth 1 https://github.com/oxy-hq/oxygenWrote 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/oxy-hq/oxygen/oxy-customer-apps-perf)<a href="https://agentmods.dev/skills/oxy-hq/oxygen/oxy-customer-apps-perf"><img src="https://agentmods.dev/badge/skills/oxy-hq/oxygen/oxy-customer-apps-perf/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/oxy-hq/oxygen/oxy-customer-apps-perf"><img src="https://agentmods.dev/badge/skills/oxy-hq/oxygen/oxy-customer-apps-perf.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 | $0.00126 | $0.02532 |
| Opus 5.5 | $0.00050 | $0.01013 |
| Sonnet 5.5 | $0.00025 | $0.00506 |
| Haiku 4.5 | $0.00013 | $0.00253 |
Grade A, and why
oxy-customer-apps-perf 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Customer-apps performance (serve + data plane)
The perf baseline every custom-app request must keep, established by PR #2634
(full rationale, threat model, and future options: internal-docs/customer-apps-performance.md).
When you touch the serve or data path, preserve these — they are cheap to keep and
expensive to retrofit under load.
Serve plane — custom_apps_serve.rs, cli/commands/serve.rs
- Content-hashed assets → immutable Cache-Control. URLs under
assets/(Vite / Astro / Rsbuild / SvelteKit) and_next/static/(Next) change only when their bytes change →public, max-age=31536000, immutable. HTML →private, no-cache; unfingerprinted root files →public, max-age=300, unless their?v=names the build being served (the launcher's icon/art URLs) → immutable. This iscache_control_for; a new hashed-asset dir means adding its prefix there. Without it, every chunk re-runs the full auth + membership walk inserve_inner. - HTML stays
private. It carries a per-visitor trackingSet-Cookie, andno-cachealone still lets a shared cache store the response — which would hand every later visitor one session id. Do not "simplify" this tono-cache. - The resolution chain must stay fully cached. Warm steady state is zero DB
round-trips per asset. Adding an uncached lookup to
serve_pretty/serve_from_s3_buildcosts ~100 queries per page load on its own. New app-row state means a newinvalidate_app_resolution_cache()call site. - Assets are pre-compressed at publish, not per request (
custom_apps_precompress). A new compressible output type belongs inis_precompressible_extension; HTML must stay excluded (it is rewritten at serve time). Any response that shipsContent-EncodingMUST also shipVary: accept-encoding. - HTML → weak ETag +
If-None-Match304. Injected HTML is a transform, so use a weakW/"…"ETag over the final bytes (etag_for/if_none_matchinserve_from_s3_build); a matchingIf-None-Matchreturns 304 and skips the re-inject. - Compression is SSE-safe.
CompressionLayeron/customer-apps/{*path}relies onDefaultPredicate, which skipstext/event-streamand already-encoded bodies — so the/fnSSE stream and the v0 proxy are untouched. If you add a streaming response under this route, confirm it staystext/event-stream(or keep it off the layer); never compress a stream.
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 · 146 lines · 126 tokens per session scan A cb51ee9c68b8
oxy-customer-apps-perf is a skill published in the GitHub repository oxy-hq/oxygen (203 stars, last pushed yesterday), licensed Apache-2.0. It adds 126 tokens to every session and 2,532 once invoked, about $0.0005 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 2026-10-14.
Other skills, from other repositories
stripe-projects
Provision SaaS services + sync creds via Stripe Projects.
django-storages-s3
Use when configuring Django to store static and media files on AWS S3 with django-storages. Invoke when working with the STORAGES setting, S3 buckets, presigned URLs, CloudFront, or boto3-backed file storage in settings.py. Configures the Django 4.2+ STORAGES dict, public/private custom backends, presigned GET/POST…
websocket-engineer
Use when building real-time communication systems with WebSockets or Socket.IO. Invoke for bidirectional messaging, horizontal scaling with Redis, presence tracking, room management.
azure-eventhub-dotnet
Azure Event Hubs SDK for .NET. Use for high-throughput event streaming: sending events (EventHubProducerClient, EventHubBufferedProducerClient), receiving events (EventProcessorClient with checkpointing), partition management, and real-time data ingestion. Triggers: "Event Hubs", "event streaming"…
azure-mgmt-botservice-dotnet
Azure Resource Manager SDK for Bot Service in .NET. Management plane operations for creating and managing Azure Bot resources, channels (Teams, DirectLine, Slack), and connection settings. Triggers: "Bot Service", "BotResource", "Azure Bot", "DirectLine channel", "Teams channel", "bot management .NET", "create bot".
wikipedia
Search and read Wikipedia via x wkp — MediaWiki API, no API key, zero install; query, extract, suggest, and DDG route in one module. Load for wiki, wikipedia, encyclopedia lookup, article summary.