Physical camera perception, measurement, and verification tool for MCP AI agents
Wellknown found it in public sources; nobody has proven control of it yet. Claiming takes one click if the repository is under your GitHub account, or a small file on your domain otherwise. Verified owners get the badge, 15-minute checks, status alerts, edits that outrank crawled data, and a ranking boost.
Agents can do it too: POST https://wellknown.network/api/v1/claims with {"agent":"agent-cam-mcp","method":"well_known_file"} — machine-readable steps at claim.json, guide at /docs/claim.
Everything here was measured by our prober or read from a registry. Nothing is self-reported.
Attributed to the source that supplied each field. Treated as claims, not facts.
# Agent Cam MCP Tool Physical camera perception, quantitative measurement, and verification tool for MCP AI agents. [](https://github.com/skr-electronics-lab/agent-cam-mcp/actions/workflows/ci.yml) [](https://pypi.org/project/agent-cam-mcp/) [](https://pypi.org/project/agent-cam-mcp/) [](https://opensource.org/licenses/MIT) [](https://modelcontextprotocol.io/) --- ## 1. Overview Agent Cam MCP Tool gives AI coding agents eyes on the physical world. It interfaces with USB webcams and RTSP/HTTP network camera streams, exposing them to any Model Context Protocol (MCP) compatible agent (Google Antigravity, Claude Code, OpenAI Codex, Cursor). ### The Problem Agents working on embedded systems, robotics, CNC machinery, or 3D printers frequently declare builds or firmware flashes "successful" while the physical hardware remains unresponsive, stuck in a boot loop, or displaying error codes. Traditional agents rely solely on build return codes and serial logs. ### The Solution Agent Cam MCP Tool allows the agent to inspect the physical hardware at any moment: - Returns images directly as MCP image content (like browser screenshots in IDE agents). Nothing is written to disk by default. - Returns quantitative numbers alongside visual frames (mean brightness, contrast, lit pixel percentage, dominant colors, edge density, sharpness, and motion level) so agents make decisions on hard evidence. - Maintains an in-memory RAM ring buffer so agents can retrieve what happened before an error or correlate visual changes with serial logs. - Includes a local developer dashboard to monitor camera feeds, define named regions o…
Mapped onto the structured taxonomy from declared text and observed tool names. Confidence shown for derived entries.
Every source is kept verbatim. Field changes are logged as events.