bitbucket-automation

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.75). SKILL.md describes runtime tool calls to Bitbucket (e.g., BITBUCKET_GET_PULL_REQUEST_DIFF, BITBUCKET_GET_PULL_REQUEST_DIFFSTAT, BITBUCKET_LIST_PULL_REQUESTS, BITBUCKET_LIST_ISSUES) whose responses can include outsider-authored free text (PR/issue comments, diffs, titles/descriptions) that the agent would read and place into LLM context.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires adding and using the MCP endpoint https://rube.app/mcp at runtime and instructs calling RUBE_SEARCH_TOOLS to fetch tool schemas which directly control the agent's available tools/behavior, so this external URL is a runtime dependency that can alter prompts/instructions.