Grok bot feels like GPT-3.5 in testing, lacks 'done receipts' — pudding protocol proposed
RachelVT42 · x · 2026-09-06
A tester found Grok bot's biggest gap is lack of verifiable 'done claims', citing the new open protocol pudding ('no pudding, no done') that makes agent completion claims carry receipts. In her testing Grok bot mostly wasted time while Astra and Grok Build actually helped her accomplish things — Astra more polished, Grok Build good enough. Grok bot remains promising for normies but feels like GPT-3.5-era frustration.
More from coding & agent
- Dev demos AI models handling 3D objects and animations surprisingly well — chongdashu · 2026-10-23
- Jason Kneen burns 104.56B tokens and $24.7K in 304 days of AI coding on Codex and Claude Code — jasonkneen · 2026-10-23
- Typosquatted Blender MCP Repo Hit 958 Stars in Days and Silently Ran Malware — Promptmethus · 2026-10-23
- Benchmarking 56 fresh dependency bugs: MCP routing lifts Claude Code accuracy 68% to 91% — Available_Side_7134 · 2026-10-23
- Swapping only the eval harness moves results 30%: designing agent evals like a scientist — AI Engineer · 2026-10-23
- Med student builds a whole-library RAG with Cognee; single-book ingestion still freezes his machine — shrike293103 · 2026-10-23