输出目录路径(可选,默认 ./output)
AI agents invoke record_video to trigger actions in Mac-Camera-MCP. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
Recording video via the system camera is an external operation that activates hardware (camera/microphone), captures potentially sensitive data, and invokes ffmpeg. This goes beyond a simple write — it triggers real-world device operations. The description is minimal (only mentions output directory), but context from the server description makes the function clear.
From the tool's definition Tool name 'record_video' on a server described as enabling 'video recording using the system camera via ffmpeg'; description mentions output directory path for recording.
Attacks that exploit this kind of access
输出目录路径(可选,默认 ./output). It is categorised as a Execute tool in the Mac-Camera-MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Mac-Camera- MCP server in PolicyLayer and add a rule for record_video: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Mac-Camera-MCP. Nothing to install.
record_video is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the record_video rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for record_video. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
record_video is provided by the Mac-Camera- MCP server (mcpservice/mac-camera-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →