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 milind-soni/axstream --skill axstreamgit clone --depth 1 https://github.com/milind-soni/axstreamWrote 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/milind-soni/axstream/axstream)<a href="https://agentmods.dev/skills/milind-soni/axstream/axstream"><img src="https://agentmods.dev/badge/skills/milind-soni/axstream/axstream/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/milind-soni/axstream/axstream"><img src="https://agentmods.dev/badge/skills/milind-soni/axstream/axstream.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Rogue Agent · line 88 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00096 | $0.03249 |
| Opus 5.5 | $0.00038 | $0.01300 |
| Sonnet 5.5 | $0.00019 | $0.00650 |
| Haiku 4.5 | $0.00010 | $0.00325 |
Grade A, and why
axstream 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 — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
axstream — replay UI tasks instead of re-deriving them
First decision: delegate multi-step UI work to the computer-use subagent
If your harness has a Task/Agent tool and an agent named computer-use
(plugin form: axstream:computer-use) is available, DELEGATE any multi-step
UI task to it instead of driving the UI from the main conversation. It runs
on a fast model with exactly the tools below, replays a matching macro when
one exists, drives novel tasks as batched verified actions, saves a macro on
success, and returns a compact outcome report — so screenshots and OCR dumps
never enter your context, and your own turns stay fast for the rest of the
session. It can run in the background while you continue other work.
- Delegate: "open X and do Y then Z", click-through flows, reproduce-a-bug in a running app, verify-something-on-screen, anything worth a macro.
- Do it yourself (tools below): a single action, or when you must reason about visual state mid-flow.
- Authorize explicitly: if the task includes an irreversible step (send,
pay, delete), name that exact step in the delegation prompt — the agent
refuses
risk:"risky"ops it wasn't explicitly given.
Fast primitives (axstream MCP tools, if connected)
When the axstream MCP server is available, prefer its primitives over
screenshot-driven computer use for mechanical steps — they run on-device in
milliseconds and return text, not images:
screen_text(~200ms) — the window as text lines with coordinates. Use INSTEAD of a screenshot to learn what's on screen.find(~30-600ms) — locate one control/text; returns a click-ready target.act— a whole batch of actions in ONE call (open/click/type/key +assertguards). Plan the sequence once; don't take a model turn per step.check(~250ms) — verify an outcome ("results page shows my query") without a verification screenshot.begin_capture+compile_capture— wrap Codex's initialized nativeskycomputer-use object, record successful actions plus their preceding AX state, then compile that real trace into a parameterized macro. This is the native fallback bridge; cua-driver's recorder cannot observeskycalls.
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 First seen · 231 lines · 96 tokens per session scan A 15c3be923188
axstream is a skill published in the GitHub repository milind-soni/axstream (20 stars, last pushed 2mo ago), licensed MIT. It adds 96 tokens to every session and 3,249 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
macos-harness
Read and operate the apps on this Mac from the shell. Use it when a task needs an app's UI and no API, CLI or purpose-built tool can do it.
chrome
Google Chrome heuristics for background-computer-use. Use when the user asks you to navigate, scroll, click links, fill forms, or otherwise drive Chrome without stealing focus. Also applies to Chromium-family browsers (Brave, Edge, Arc) that share the same AppleScript JavaScript execution primitive.
notion
Notion-specific heuristics for driving the Notion desktop app via background-computer-use. Use when the user asks you to edit a Notion page, add a block, create a database row, or otherwise interact with Notion's UI.
safari
Safari desktop heuristics for background-computer-use. Use when the user asks you to navigate, scroll, click links, switch tabs, or otherwise drive Safari without stealing focus.
clock
Clock.app heuristics for background-computer-use. Use when the user asks you to set an alarm, start a timer, run a stopwatch, or manage world clocks.
computer-use
Read and drive native desktop apps through the accessibility layer — list apps, snapshot a window as a numbered element tree, then click/type/scroll/drag by element index or screen coordinates. Use for a desktop app, not a web page. macOS + Windows; off unless enabled in Settings.