全文搜索文档条目，返回摘要和匹配片段（不含全文）。
AI agents call project_doc_search to retrieve information from ProjectDoc without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool only retrieves and filters existing documentation based on search criteria. It returns summarized results and matching text fragments without creating, modifying, deleting, or executing any operations.
From the tool's definition Tool description states '全文搜索文档条目，返回摘要和匹配片段' which translates to 'full-text search of document entries, returns summaries and matching snippets'. The search operation retrieves and queries data without modification or side effects.
Attacks that exploit this kind of access
全文搜索文档条目，返回摘要和匹配片段（不含全文）。. It is categorised as a Read tool in the ProjectDoc MCP Server, which means it retrieves data without modifying state.
Register the ProjectDoc MCP server in PolicyLayer and add a rule for project_doc_search: 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 ProjectDoc. Nothing to install.
project_doc_search 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 project_doc_search 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 project_doc_search. 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.
project_doc_search is provided by the ProjectDoc MCP server (yarrow-cai/project-doc). 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 →