{"version":"2.1.311","anchor":"mcp-image-results-can-be-rewritten-by-a-readpicture-hook-pe","canonical_anchor":"mcp-image-results-can-be-rewritten-by-a-readpicture-hook-pe","heading":"Permission requests now say which subagent asked","tier":"notice","area":"Permissions","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.311\/e\/mcp-image-results-can-be-rewritten-by-a-readpicture-hook-pe","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.311","markdown":"### Permission requests now say which subagent asked\n\nTool permission requests now carry an `agent_id` for the subagent that asked, and one scan skips subagent requests; MCP files gain a `readPicture` step\n\n**Unclear.** It is not clear which MCP results go through the picture step or what it changes about them.\n\n**What**\n\nA permission request is the question Claude Code asks before it uses a tool. These requests now include an `agent_id`, taken from `askerAgentId`. It identifies the subagent (a helper agent Claude starts for part of a task) that made the request.\n\n- Tool permission requests now include `agent_id`.\n\n- One loop that goes through pending permission requests now skips any request with a non-empty `agent_id`, so requests from subagents are not handled there.\n\n- Separately, non-image results from MCP servers (outside services that give Claude extra tools) that are saved to a file now pass through `readPicture`.\n\n**Why**\n\nPrograms that show or answer permission prompts from outside the terminal, such as remote approval screens, can now tell which subagent is asking. This may also change which pending prompts are shown or replayed.\n\n- Area: Permissions\n- Names: `agent_id`\n- Tier: You'll notice\n- Useful: 2\/5\n- Signal: 2\/5\n- Scope: individual\n- Heads-up: no"}