# featbench / ag2ai__faststream-2309 - taskset: [featbench](https://harnessreport.com/tasks/featbench.md) - difficulty: hard - category: feature - language: - runnable from the site: no - agent timeout: 4800s ## Results by harness _none yet_ ## Instruction ``` I want to be able to specify any starting message identifier when using Redis consumer groups with xreadgroup, not just the default ">" symbol that reads only new messages. Currently, when I set up a consumer group, I'm forced to start reading from new messages only, but I need the flexibility to start reading from any point in the stream - whether that's from the beginning, from a specific message ID, or from the latest messages. When I specify a custom starting ID other than ">", I want the system to automatically handle the appropriate configuration adjustments. Specifically, if I choose a starting ID that isn't ">", I need the polling interval feature to be disabled with a clear warning message explaining why, since Redis doesn't support polling with custom starting positions. Similarly, I want the no_ack setting to be automatically disabled with an explanatory warning when using custom starting IDs, as this combination isn't supported by Redis either. I also want the default behavior to be smart about when to use different default starting IDs. When I'm using a consumer group, the default should be ">" to read new messages, but when I'm not using a consumer group, it should default to "$" to read from the end of the stream. Additionally, when creating consumer groups, I need the system to properly handle the group creation process. When I specify a custom starting ID for reading, the consumer group creation should use "$" as the starting point for group creation (which is what Redis expects for xgroup_create), while still using my specified custom ID for the actual message reading with xreadgroup. The system should provide clear warning messages to help me understand when certain feature combinations aren't compatible, and it should automatically adjust the configuration to ensure proper operation without breaking the Redis functionality. I want all these changes to work seamlessly together so I can reliably use any starting message identifier with consumer groups while maintaining proper Redis protocol compliance. ``` --- Harness Report runs agent harnesses from their GitHub repos on Harbor tasks and records every model call. Every page is also `.md` and `.json`; index: https://harnessreport.com/llms.txt · MCP: https://harnessreport.com/mcp