キーワードでコンテキストを検索します
AI agents call search_contexts to retrieve information from OCD - Organized Context Datastore (MCP) without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool performs a search operation to retrieve matching contexts from the datastore. This is a read-only query operation with no side effects, creation, modification, deletion, code execution, or financial impact. The search functionality returns matching results but does not alter any stored data.
From the tool's definition Tool name 'search_contexts' and description 'キーワードでコンテキストを検索します' (search contexts by keyword) indicate querying/retrieving stored context data without modification.
Attacks that exploit this kind of access
キーワードでコンテキストを検索します. It is categorised as a Read tool in the OCD - Organized Context Datastore (MCP) MCP Server, which means it retrieves data without modifying state.
Register the OCD - Organized Context Datastore (MCP) MCP server in PolicyLayer and add a rule for search_contexts: 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 OCD - Organized Context Datastore (MCP). Nothing to install.
search_contexts 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 search_contexts 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 search_contexts. 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.
search_contexts is provided by the OCD - Organized Context Datastore (MCP) MCP server (teamstove/organized-context-datastore-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 →