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 instructions/huseyinstif/oculos/agents-mdgit clone --depth 1 https://github.com/huseyinstif/oculosWrote 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/instructions/huseyinstif/oculos/agents-md)<a href="https://agentmods.dev/instructions/huseyinstif/oculos/agents-md"><img src="https://agentmods.dev/badge/instructions/huseyinstif/oculos/agents-md.svg" alt="Measured on agentmods" 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 | $0.01678 | $0.01678 |
| Opus 5 | $0.00839 | $0.00839 |
| Sonnet 5 | $0.00336 | $0.00336 |
| Haiku 4.5 | $0.00168 | $0.00168 |
Grade A, and why
oculos AGENTS.md 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 4d 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 — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OculOS — AI Agent Instructions
You have access to OculOS, a local REST API that lets you read and control any desktop application through its UI Automation tree.
Base URL: http://127.0.0.1:7878
Core Workflow
1. list_windows → find the target application (get pid / hwnd)
2. find_elements / tree → locate the elements you need (get oculos_id)
3. interact → act on those elements
Important:
oculos_idvalues are session-scoped. They're valid after the tree is built but may change if you call tree/find again. Always fetch the element immediately before interacting with it.
Discovery
List all windows
GET /windows
Returns: [{ pid, hwnd, title, exe_name, rect, visible }]
Search for elements (preferred — fast)
GET /windows/{pid}/find?q=Submit&type=Button&interactive=true
GET /hwnd/{hwnd}/find?q=Search&interactive=true
Parameters:
q— case-insensitive substring match on label or automation_idtype— element type filter (Button, Edit, CheckBox, ListItem, ComboBox…)interactive=true— return only elements that have at least one action
Full UI tree (for exploration)
GET /windows/{pid}/tree
GET /hwnd/{hwnd}/tree
Use the HWND variant when a process has multiple windows (e.g. Teams).
Interactions
All interaction endpoints accept a JSON body where noted.
| Endpoint | Body | When to use |
|---|---|---|
POST /interact/{id}/click |
— | Buttons, links, menu items |
POST /interact/{id}/set-text |
{"text":"…"} |
Input fields (replaces all text) |
POST /interact/{id}/send-keys |
{"keys":"…"} |
Keyboard simulation (see below) |
POST /interact/{id}/focus |
— | Move keyboard focus |
POST /interact/{id}/toggle |
— | CheckBox, ToggleButton |
POST /interact/{id}/expand |
— | ComboBox, TreeItem, MenuItem |
POST /interact/{id}/collapse |
— | ComboBox, TreeItem, MenuItem |
POST /interact/{id}/select |
— | ListItem, RadioButton, TabItem |
POST /interact/{id}/set-range |
{"value":75} |
Slider, Spinner |
POST /interact/{id}/scroll |
{"direction":"down"} |
Scroll containers |
POST /interact/{id}/scroll-into-view |
— | Bring element into viewport |
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.
- 4d ago First seen · 214 lines · 1,678 tokens per session scan A da1b91f3aec3
oculos AGENTS.md is an instructions file published in the GitHub repository huseyinstif/oculos (130 stars, last pushed 6mo ago), licensed MIT. It adds 1,678 tokens to every session, about $0.0084 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.
Other instructions, from other repositories
Orbination-AI-Desktop-Vision-Control CLAUDE.md
Instructions for amichail-1/Orbination-AI-Desktop-Vision-Control, covering desktop control mcp - ai usage guide, observation priority (how to see the screen), action priority (how to interact), workflow pattern and anti-patterns (do not).
agent-desktop CLAUDE.md
Claude Code instructions for lahfir/agent-desktop, covering claude.md, common commands, pre-commit hook, project overview and git & commits.
Shinken CLAUDE.md
Instructions for Meirtz/Shinken, covering claude.md — guide for ai coding sessions in this repo, ⛔ the one hard rule: this is a public open-source project, layout, development commands and architecture (how the pieces connect).
kwin-mcp CLAUDE.md
Instructions for isac322/kwin-mcp, covering claude code working guidelines, project overview, toolchain, common commands and code style.
trope-cua AGENTS.md
Instructions for tropeai/trope-cua, covering repository guidelines, collaboration & git safety, repository layout, quick commands and engineering guidelines.
mcp-server-uiautomation copilot-instructions.md
Copilot instructions for trsdn/mcp-server-uiautomation, covering github copilot instructions - uiautomationmcp, project focus, read first, core rules and quick validation.