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/uilibs/uiprotect/agents-mdgit clone --depth 1 https://github.com/uilibs/uiprotectWrote 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/uilibs/uiprotect/agents-md)<a href="https://agentmods.dev/instructions/uilibs/uiprotect/agents-md"><img src="https://agentmods.dev/badge/instructions/uilibs/uiprotect/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 | $0.06170 | $0.06170 |
| Opus 5 | $0.03085 | $0.03085 |
| Sonnet 5 | $0.01234 | $0.01234 |
| Haiku 4.5 | $0.00617 | $0.00617 |
Grade A, and why
uiprotect 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 — 460 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Notes for LLM contributors
A short orientation file for an LLM working in this repo. Skim before making changes; keep edits consistent with what's described here. Read README.md for the user-facing intro and CONTRIBUTING.md for the human-facing contributor guide.
What this project is
uiprotect is an unofficial async Python client + CLI for the
UniFi Protect surveillance NVR. It is the library that the Home
Assistant unifiprotect integration is built on top of, and it
is the de-facto Python binding for talking to a UniFi Protect
console — Ubiquiti does not publish an official one. Public API
is exported from the top-level uiprotect package; the main
entry point is uiprotect.ProtectApiClient in src/uiprotect/api.py.
There is no upstream protocol owner. The wire format is whatever
the UniFi Protect server in front of you serves: a partially
documented REST API plus a binary WebSocket update stream. Both
shift across firmware releases, and the test corpus under
tests/sample_data/ is the closest thing to a reference. Behaviour
changes that depend on a specific Protect firmware version should
say so in the commit message.
The project was forked from pyunifiprotect after that project
relicensed away from MIT; staying MIT-licensed is a hard
requirement so the Home Assistant core integration can keep
depending on it.
Code style
- Docstrings: terse, default to single-line. A docstring is
the function's contract, not its narrative. Almost every
docstring should be one line —
"""Summary."""— describing what the function does and what the caller can pass. Multi-line is the exception, only justified when there is non-obvious caller-visible behaviour the type signature and parameter names don't already convey. Ruff is configured with most of theDfamily disabled (D100–D107,D2xx,D4xx) precisely so that missing docstrings on internal helpers don't flag — only add one when it earns its keep.
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 · 460 lines · 6,170 tokens per session scan A 186ec49e3321
uiprotect AGENTS.md is an instructions file published in the GitHub repository uilibs/uiprotect (101 stars, last pushed 5d ago), licensed MIT. It adds 6,170 tokens to every session, about $0.0309 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
pypylon AGENTS.md
Instructions for basler/pypylon, covering agents.md, repository map, environment, common commands and or, to (re)build in place after changing bindings.
aiohomematic CLAUDE.md
Instructions for SukramJ/aiohomematic, covering claude.md - ai assistant guide for aiohomematic, ⚠️ critical: cuxd/ccu-jack special handling, what makes cuxd/ccu-jack different, key constants to check and in aiohomematic/const.py.
aiohomematic copilot-instructions.md
Instructions for SukramJ/aiohomematic, covering github copilot instructions for aiohomematic, project overview, mandatory code standards, required import and type annotations.
CubismExternalEditMCP CLAUDE.md
Instructions for nana7chi/CubismExternalEditMCP, covering claude.md, 构建 / 运行 / 测试, 安装开发依赖（在项目根目录）, 语法检查 and 导入检查.
kvm-pilot CLAUDE.md
Instructions for DustinTrap/kvm-pilot, covering claude.md, what this is, non-negotiable conventions, engineering principles (how to make changes) and layout.
CarolinaCodeConferenceBadge2026 AGENTS.md
Instructions for circuitboardmedics/CarolinaCodeConferenceBadge2026, covering agents.md — circuitpython drive guidance, what this directory is, execution model, hardware — esp32-s3-wroom-1-n8 and display — hs180s10b (st7735s, 160×128, 1.77 inch).