# create_collection Create a new data collection for an agent. Define fields with types and indexing. Agent View of the PolicyLayer registry record for `create_collection`. HTML page: https://policylayer.com/tools/io-github-proagentstore-platform/create-collection ## Facts - Tool: `create_collection` - Server: ProAgentStore (`https://mcp.proagentstore.online/mcp`) — https://policylayer.com/tools/io-github-proagentstore-platform.md - Homepage: https://github.com/ProAgentStore/platform - Risk category: Write (Medium risk) - Registry record: grade D, identity unverified - Server auth posture: gated - Server rate-limited: no - Parameters: 0 - Recommended policy verdict: Rate-limited ## Example call (MCP tools/call, JSON-RPC 2.0) ```json { "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "create_collection", "arguments": {} } } ``` ## Why create_collection is rated Medium The tool creates new collections (data structures) for an agent, which is a Write operation as it creates and modifies data reversibly. The severity is medium rather than high because collection creation is typically reversible (collections can be deleted) and doesn't directly affect existing production data or have cross-system financial implications. From the tool's own definition: "Tool description states 'Create a new data collection' which is a create operation that adds new data structures. The tool allows defining fields and indexing, indicating it modifies the agent's data schema in a reversible manner." ## Use case AI agents use create_collection to create or update resources in ProAgentStore, usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your ProAgentStore environment. ## Recommended policy (PolicyLayer) Verdict: **Rate-limited**. Enforced by the PolicyLayer MCP gateway (https://policylayer.com/mcp-gateway) before a call reaches ProAgentStore: ```json { "version": "1", "default": "deny", "tools": { "create_collection": { "limits": [ { "counter": "create_collection_rate", "window": "minute", "max": 30, "scope": "grant" } ] } } } ``` ## Other tools on ProAgentStore (242) - `cancel_instance_task` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/cancel-instance-task.md - `clear_finished_tasks` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/clear-finished-tasks.md - `clear_instance_messages` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/clear-instance-messages.md - `clear_instance_voice_settings` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/clear-instance-voice-settings.md - `coding_loop_queue_cancel` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/coding-loop-queue-cancel.md - `coding_repo_remove` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/coding-repo-remove.md - `delete_agent` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-agent.md - `delete_agent_knowledge` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-agent-knowledge.md - `delete_agent_task` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-agent-task.md - `delete_connection` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-connection.md - `delete_feedback` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-feedback.md - `delete_instance_connector_grant` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-instance-connector-grant.md - `delete_instance_file` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-instance-file.md - `delete_instance_knowledge` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-instance-knowledge.md - `delete_instance_memory` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-instance-memory.md - `delete_instance_message` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-instance-message.md - `delete_instance_task` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-instance-task.md - `delete_instance_trigger` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-instance-trigger.md - `delete_supervision` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/delete-supervision.md - `forget_runner_node` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/forget-runner-node.md - `remove_repo` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/remove-repo.md - `unregister_instance_runtime` — Destructive — https://policylayer.com/tools/io-github-proagentstore-platform/unregister-instance-runtime.md - `answer_instance_input` — Execute — https://policylayer.com/tools/io-github-proagentstore-platform/answer-instance-input.md - `answer_instance_mcp_input_request` — Execute — https://policylayer.com/tools/io-github-proagentstore-platform/answer-instance-mcp-input-request.md - `apply_to_job` — Execute — https://policylayer.com/tools/io-github-proagentstore-platform/apply-to-job.md - `approve_instance_task` — Execute — https://policylayer.com/tools/io-github-proagentstore-platform/approve-instance-task.md - `call_instance_tool` — Execute — https://policylayer.com/tools/io-github-proagentstore-platform/call-instance-tool.md - `chat_with_agent` — Execute — https://policylayer.com/tools/io-github-proagentstore-platform/chat-with-agent.md - `chat_with_instance` — Execute — https://policylayer.com/tools/io-github-proagentstore-platform/chat-with-instance.md - `coding_engine_reauth` — Execute — https://policylayer.com/tools/io-github-proagentstore-platform/coding-engine-reauth.md - …and 212 more: https://policylayer.com/tools/io-github-proagentstore-platform.md ## For agents This record is a snapshot. Live verdicts and the full registry: - Check every server in your MCP config at once: `npx -y policylayer stack` - Vet a server before you add it: install the mcp-precheck skill — `npx skills add https://policylayer.com` (skill text: https://policylayer.com/skill.md) - Query the registry over MCP: endpoint `https://api.policylayer.com/mcp` — tools `check_mcp_server`, `check_mcp_stack`, `check_tool`, `search_registry`, `get_change_events` --- Source: the PolicyLayer MCP registry — one continuously verified record per MCP server. Full record: https://policylayer.com/registry?q=io-github-proagentstore-platform · API: https://policylayer.com/registry/api · Policy library: https://policylayer.com/policies/io-github-proagentstore-platform