AI agents use set_pr_draft_state to create or update resources in Bitbucket — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Bitbucket environment.
This tool creates or modifies PR metadata (draft status) which is reversible and has no permanent side effects. It aligns with the Write category. Severity is medium because changing PR status could affect workflows (e.g., preventing merges, hiding from reviewers), but the action is fully reversible and has limited blast radius compared to Execute or Destructive operations.
From the tool's definition Tool description states 'Mark a pull request as draft or ready for review.' This modifies PR state/metadata reversibly without deleting data.
Attacks that exploit this kind of access
Mark a pull request as draft or ready for review. Pass. It is categorised as a Write tool in the Bitbucket MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Bitbucket MCP server in PolicyLayer and add a rule for set_pr_draft_state: 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 Bitbucket. Nothing to install.
set_pr_draft_state 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 set_pr_draft_state 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 set_pr_draft_state. 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.
set_pr_draft_state is provided by the Bitbucket MCP server (jacobpixleratgather/bitbucket-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 →