删除一个已 ingest 的原始文件，并自动处理相关 wiki 页面。
AI agents call delete_source to permanently remove resources in Llm Wiki — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
The tool permanently removes data (source files and associated wiki pages) without the ability to undo. This meets the Destructive category definition of irreversibly deleting data. Severity is high because accidental deletion of wiki sources could result in significant information loss, though the blast radius is contained to a specific wiki system rather than critical infrastructure.
From the tool's definition Tool name is 'delete_source' and description states it deletes ('删除') an ingested source file and automatically handles related wiki pages. This is an irreversible deletion operation.
Attacks that exploit this kind of access
删除一个已 ingest 的原始文件，并自动处理相关 wiki 页面。. It is categorised as a Destructive tool in the Llm Wiki MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Llm Wiki MCP server in PolicyLayer and add a rule for delete_source: 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 Llm Wiki. Nothing to install.
delete_source 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 delete_source 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 delete_source. 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.
delete_source is provided by the Llm Wiki MCP server (quick-sort/llm-wiki-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 →