What is mata-kadalz AGENTS.md?

Repository guidance for AI agents working on mata-kadalz, a local image-understanding server that connects an MCP client to a llama-server running the Qwen3-VL-4B model.

Use it when modifying the server, its one image-inspection tool, caching, queue limits, path checks, or connection to llama-server.

It records the project rules and design limits, so changes do not accidentally alter the validated model, file-access rules, caching, or request handling.

How kadalzbaiq/mata-kadalz configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to kadalzbaiq/mata-kadalz. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/kadalzbaiq/mata-kadalz/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/kadalzbaiq/mata-kadalz

Made for: Codex, OpenCode.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/kadalzbaiq/mata-kadalz/agents-md/github.svg)](https://agentmods.dev/instructions/kadalzbaiq/mata-kadalz/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/kadalzbaiq/mata-kadalz/agents-md"><img src="https://agentmods.dev/badge/instructions/kadalzbaiq/mata-kadalz/agents-md/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.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/kadalzbaiq/mata-kadalz/agents-md"><img src="https://agentmods.dev/badge/instructions/kadalzbaiq/mata-kadalz/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 710 This file is loaded in full into every session.

When invoked 710 The same file — it is already loaded in full.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00710	$0.00710
Opus 5	$0.00355	$0.00355
Sonnet 5	$0.00142	$0.00142
Haiku 4.5	$0.00071	$0.00071

Measured 10d ago against content hash 123adf16035b, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

mata-kadalz AGENTS.md scanned grade A with 1 finding 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 10d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Single-file server: `server.py`. stdlib only (`urllib`, `asyncio`, `hashlib`,

AGENTS.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Guide for AI agents working in this repo.

What this is

A local vision MCP server for any MCP client. mata-kadalz (Python, single file server.py) talks to a llama-server HTTP endpoint (native on Windows, Linux, macOS, or in WSL) that runs Qwen3-VL-4B.

Golden rules

Never change the model. Qwen3VL-4B-Instruct-Q4_K_M + mmproj F16 is the validated combination. Model/quant changes require re-validation.
One tool only: vision.inspect(image_path, task).
Cache contract: key = sha256(image_sha256 + task + model_id); failed requests never cached; inference serialized via asyncio.Lock.
Backpressure: queue bounded by VISION_MAX_QUEUE (default 4); full queue returns LLAMA_BUSY immediately.
File access: image_path resolves against the server machine; if VISION_IMAGE_ROOTS is set, only paths inside a root (after resolve()) are allowed, else IMAGE_PATH_NOT_ALLOWED. Default (empty) = any path.
Never return image bytes — text JSON only.
Dependency boundary: mata-kadalz is the MCP layer only. Never vendor, bundle, download, or manage llama.cpp/model files in this repo. The docs point users to official llama.cpp/Hugging Face sources; the package only connects to an already-running llama-server.

Architecture

client -> vision.inspect -> mata-kadalz server.py (stdio or streamable HTTP)
                             -> HTTP POST http://<llama-server>:9931/v1/chat/completions
                             -> llama-server (Qwen3-VL-4B GGUF + mmproj)

Conventions

Single-file server: server.py. stdlib only (urllib, asyncio, hashlib, mimetypes) plus the mcp SDK; uvicorn used only for --transport http.
Config precedence: DEFAULTS < config/config.json (empty skipped) < env.
Relative config paths resolve against repo root (BASE_DIR).
Host detection: 127.0.0.1 unless WSL, then gateway IP; LLAMA_SERVER_URL wins.
Tests: pytest in tests/, must not require llama-server running.
Logging: file handler into <runtime>/vision/logs/vision-mcp.log.
Commit style: Conventional Commits, matching git log.

Read the full file on GitHub · 61 lines

Changes

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.

10d ago First seen · 61 lines · 710 tokens per session scan A 123adf16035b

Subscribe to this mod's changes

mata-kadalz AGENTS.md is an instructions file published in the GitHub repository kadalzbaiq/mata-kadalz (0 stars, last pushed 2d ago), licensed MIT. It adds 710 tokens to every session, about $0.0036 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Python llama-cpp llama-server mcp model-context-protocol qwen3-vl

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

mata-kadalz: Instructions file for Codex