Export a collection to JSON for backup or transfer to another system.
AI agents call export_collection to retrieve information from ChromaDB Local MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
Exporting a collection reads data from ChromaDB and serializes it to JSON. It does not modify, delete, or create data within the database — it is a read/retrieval operation. Severity is medium because the export could expose large volumes of potentially sensitive vector data and metadata to an external destination or file, representing a data-exfiltration risk if misused.
From the tool's definition Export a collection to JSON for backup or transfer to another system
Attacks that exploit this kind of access
Export a collection to JSON for backup or transfer to another system. It is categorised as a Read tool in the ChromaDB Local MCP Server MCP Server, which means it retrieves data without modifying state.
Register the ChromaDB Local MCP Server MCP server in PolicyLayer and add a rule for export_collection: 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 ChromaDB Local MCP Server. Nothing to install.
export_collection 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 export_collection 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 export_collection. 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.
export_collection is provided by the ChromaDB Local MCP Server MCP server (vespo92/chromadblocal-mcp-server). 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 →