Watch for new annotations on a localhost project. This tool blocks until pending annotations appear, then returns them. Use this in a loop for hands-free mode: call watch_annotations → implement each annotation → call delete_annotation → call watch_annotations again. The tool polls every 10 secon...
AI agents call watch_annotations to retrieve information from Vibe Annotations Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Even though watch_annotations only reads data, uncontrolled read access leaks sensitive information and racks up API costs — an agent caught in a retry loop can make thousands of calls a minute without anyone noticing.
Risk signalsBulk/mass operation — affects multiple targets
Attacks that exploit this kind of access
Watch for new annotations on a localhost project. This tool blocks until pending annotations appear, then returns them. Use this in a loop for hands-free mode: call watch_annotations → implement each annotation → call delete_annotation → call watch_annotations again. The tool polls every 10 seconds and automatically stops after the timeout period (default 5 minutes) of no new annotations appearing. IMPORTANT: You must know the localhost URL of the project you are watching. If you do not know it, ask the user before calling this tool. Example:. It is categorised as a Read tool in the Vibe Annotations Server MCP Server, which means it retrieves data without modifying state.
Register the Vibe Annotations Server MCP server in PolicyLayer and add a rule for watch_annotations: 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 Vibe Annotations Server. Nothing to install.
watch_annotations is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the watch_annotations 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 watch_annotations. 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.
watch_annotations is provided by the Vibe Annotations Server MCP server (vibe-annotations-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.