mcp_camera AGENTS.md

mcp_camera AGENTS.md is an instructions file for Codex, OpenCode from bkosciow/mcp_camera. It costs 1,629 tokens per session, scanned B, original, MIT.

Project instructions for a camera service that captures fresh images from attached USB cameras and exposes them through an HTTP API, with optional agent tools.

In plain words
What is it for?
Use them when developing, testing, linting, or type-checking the camera service and its optional MCP integration.
Why use it?
They give an agent the project’s structure, technology choices, testing approach, and camera-handling rules while it works on the code.

Instructions file for CodexOpenCode

Install

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.

agentmods
npx agentmods add instructions/bkosciow/mcp_camera/agents-md
Clone the repo
git clone --depth 1 https://github.com/bkosciow/mcp_camera

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.

agentmods badge for mcp_camera AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/bkosciow/mcp_camera/agents-md.svg)](https://agentmods.dev/instructions/bkosciow/mcp_camera/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/bkosciow/mcp_camera/agents-md"><img src="https://agentmods.dev/badge/instructions/bkosciow/mcp_camera/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,629 This file is loaded in full into every session.
When invoked 1,629 The same file — it is already loaded in full.
Security scan B 1 finding. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.01629 $0.01629
Opus 5 $0.00814 $0.00814
Sonnet 5 $0.00326 $0.00326
Haiku 4.5 $0.00163 $0.00163

Measured 2d ago against content hash 75ec8c3241ef, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

mcp_camera AGENTS.md scanned grade B 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 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.

Reads MCP configurationmediumAgent snooping

mcp.json carries server URLs and auth tokens; reading it lets a mod discover and abuse other integrations.

Inbound requests to the MCP server itself are also protected on network transports: clients must send `Authorization: Bearer <MCP_AUTH_TOKEN>` on every request (no query-parameter fallback), and streamable-http/sse refus
AGENTS.md · 147 lines

How it starts

The opening of the file, as written. The whole thing — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Camera MCP — Agent Instructions

Project Overview

USB camera snapshot service that captures fresh images from an attached USB camera and serves them via a simple HTTP API. An optional MCP server exposes camera tools to Claude Code for live image capture.

Core value: Every API call captures a new frame — no stale caches, no pre-recorded footage.

Tech Stack

  • Python 3.12+ with uv (package manager)
  • FastAPI for the HTTP layer
  • OpenCV (cv2) for camera capture and image processing
  • uvicorn ASGI server
  • pydantic-settings for configuration
  • pytest for testing (all tests mock cv2.VideoCapture — no hardware needed)
  • ruff for linting, mypy for type checking

Architecture

Single-container stateless service with three components:

1. Camera Manager (src/camera_mcp/camera.py)

Singleton that manages multiple USB cameras. Scans /dev/video* devices first, then falls back to indices 0–5. Configures highest native resolution (1920×1080 → 1280×720 fallback). Discards stale buffered frames before each capture for freshness. Provides indexed access via get(i) and __getitem__(i).

2. HTTP API (src/camera_mcp/main.py)

FastAPI application with three endpoints:

Endpoint Method Description
/capture GET Capture from first camera (index 0)
/capture/{cam_index} GET Capture from a specific camera by index
/camera GET Info for all detected cameras
/camera/{cam_index} GET Info for a specific camera by index
/health GET Service health, location name(s), all camera states, uptime, last error

All endpoints except /health require the header Authorization: Bearer <CAMERA_AUTH_TOKEN>.

3. MCP Server (src/camera_mcp/mcp_server.py)

Model Context Protocol server exposing two tools for Claude Code:

  • capture_image(camera_index, max_width) — calls /capture/{cam_index} (or /capture for index 0), returns the JPEG as an Image object. camera_index is 0-based, defaults to 0.
  • camera_status() — calls /health, returns formatted status string for all detected cameras, including the deployment's location.

Read the full file on GitHub · 147 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.

  1. 2d ago First seen · 147 lines · 1,629 tokens per session scan B 75ec8c3241ef

Subscribe to this mod's changes

mcp_camera AGENTS.md is an instructions file published in the GitHub repository bkosciow/mcp_camera (0 stars, last pushed 12d ago), licensed MIT. It adds 1,629 tokens to every session, about $0.0081 per session on Opus 5. A static security scan graded it B with 1 finding (reads mcp configuration). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other instructions, from other repositories

brilliant_sdk AGENTS.md

Instructions for brilliantlabsAR/brilliant_sdk, covering brilliant sdk — agent guide, how an app works (the pattern behind everything), minimal reading paths, verify without hardware and testing.

brilliantlabsAR/brilliant_sdk · 1,817 tokens

NeoMind CLAUDE.md

Instructions for camthink-ai/NeoMind, covering neomind — edge ai platform for iot, development commands, ecosystem repositories, extension package contract (.nep) and device type template contract (json).

camthink-ai/NeoMind · 4,296 tokens

AMD-BC-250-Windows-Driver AGENTS.md

Instructions for Keshas-dev/AMD-BC-250-Windows-Driver, covering amd bc-250 windows driver — agent notes, host machine, ⚠️ historical-data caution (2026-08-21, strengthened), fresh re-verifications (current build, 2026-08-21) and workspace boundary.

Keshas-dev/AMD-BC-250-Windows-Driver · 48,206 tokens

Nothan-OS CLAUDE.md

Instructions for Vinalinux-Org/Nothan-OS, covering claude.md — nothanos, hard rules, toolchain, driver and design philosophy.

Vinalinux-Org/Nothan-OS · 499 tokens

zmk-config AGENTS.md

AGENTS.md instructions for urob/zmk-config, covering customization guide, ground rules, how the multi-board layout works, adding a new board and where to change what.

urob/zmk-config · 1,867 tokens

ext-apps CLAUDE.md

Claude Code instructions for modelcontextprotocol/ext-apps, a project described as: Official repo for spec & SDK of MCP Apps protocol - standard for UIs embedded AI chatbots, served by MCP servers.

modelcontextprotocol/ext-apps · 3 tokens