AI agents call compare to retrieve information from Yunxiao without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
| Parameter | Type | Required | Description |
|---|---|---|---|
to | string | Yes | Target commit SHA, branch, or tag. |
from | string | Yes | Source commit SHA, branch, or tag. |
straight | string | — | Whether to compare directly without merge base. |
sourceType | string | — | Source ref type: branch, tag, or commit. |
targetType | string | — | Target ref type: branch, tag, or commit. |
repositoryId | string | Yes | Repository numeric ID or full path such as org/repo. |
organizationId | string | — | Yunxiao organization ID. When omitted, the server uses the user's default organization. |
Parameters from the server's own tool schema.
This tool queries and displays differences between repository versions (commits, branches, tags) but does not execute code, modify data, delete anything, or trigger deployments. It is a passive read operation similar to 'git diff'. Blast radius is minimal — an AI agent cannot cause harm by comparing commits.
From the tool's definition Tool description states it 'Returns the diff between the two refs' — a retrieval and comparison operation with no data modification or execution. The verb 'compare' and 'returns diff' indicate read-only analysis of repository state.
Attacks that exploit this kind of access
Compare two commits, branches, or tags in a CodeUp repository. Returns the diff between the two refs. It is categorised as a Read tool in the Yunxiao MCP Server, which means it retrieves data without modifying state.
compare accepts 7 parameters: to, from, straight, sourceType, targetType, repositoryId, organizationId. Required: to, from, repositoryId. The full parameter table on this page comes from the server's own tool schema.
Register the Yunxiao MCP server in PolicyLayer and add a rule for compare: 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 Yunxiao. Nothing to install.
compare 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 compare 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 compare. 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.
compare is provided by the Yunxiao MCP server (@futuretea/yunxiao-mcp-server). 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 →