Create a KB article (draft). Set publish true with collectionId and isPublic to publish in one step
AI agents use create_article to create or update resources in SparrowDesk — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your SparrowDesk environment.
This tool creates new knowledge base articles, which is a Write operation that modifies the system's data store by adding new content. While reversible through deletion, the creation of public-facing KB articles could spread misinformation or inappropriate content if an AI agent misuses the parameters.
From the tool's definition Tool name is 'create_article' and description states 'Create a KB article (draft)' which indicates content creation.
Attacks that exploit this kind of access
Create a KB article (draft). Set publish true with collectionId and isPublic to publish in one step. It is categorised as a Write tool in the SparrowDesk MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the SparrowDesk MCP server in PolicyLayer and add a rule for create_article: 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 SparrowDesk. Nothing to install.
create_article is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the create_article 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 create_article. 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.
create_article is provided by the SparrowDesk MCP server (sparrowdesk/sparrowdesk-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 →