Batch operations on AgentDB episodes (insert, update, delete). Note: entries are stored in the AgentDB episodes table, not the memory_search namespace. Use memory_store for entries that should be searchable via memory_search. Use when generic memory_* tools are wrong because you need AgentDB-spec...
AI agents call agentdb_batch to permanently remove resources in Claude Flow — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
The tool explicitly supports 'delete' operations in batch mode, which can irreversibly remove AgentDB episode entries. Since delete is included among the batch operations, the most severe applicable category is Destructive. Batch delete amplifies the blast radius significantly — an AI agent could accidentally or maliciously wipe large numbers of episode records in a single operation.
From the tool's definition Batch operations on AgentDB episodes (insert, update, delete)
Attacks that exploit this kind of access
Batch operations on AgentDB episodes (insert, update, delete). Note: entries are stored in the AgentDB episodes table, not the memory_search namespace. Use memory_store for entries that should be searchable via memory_search. Use when generic memory_* tools are wrong because you need AgentDB-specific controllers (HNSW vector search, hierarchical tiers, causal-graph links, pattern store/recall, RaBitQ quantization). For simple key-value persistence, memory_store/memory_retrieve are simpler. For unrelated file work, native Read/Write are fine. It is categorised as a Destructive tool in the Claude Flow MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Claude Flow MCP server in PolicyLayer and add a rule for agentdb_batch: 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 Claude Flow. Nothing to install.
agentdb_batch is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the agentdb_batch 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 agentdb_batch. 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.
agentdb_batch is provided by the Claude Flow MCP server (claude-flow). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.