A "standing review agent" seems to be one of the main differences beyond the new connectors and in place visualisation tools.
>A standing reviewer agent. This runs in the background during a session, checking citations against sources, flagging numbers it can't trace back to evidence, and catching figures that don't match the code that supposedly generated them. That's not something Code or Cowork do automatically — you'd have to ask Claude to double-check itself as a separate step.
The Netherlands is almost legendary for its agricultural productivity. Its greenhouse operations were the model for NZ capsicum production and other efforts. It also leads food science research in some areas. Wageningen is perhaps the best in that field.
A friend of mine has worked in TV and film for decades. Many times he has told me about rumoured offenders (typically after they are arrested), but other than avoiding working on productions with them what are his choices? Trying to do a completely ridiculous "citizen's arrest"?
For those interested in radioligand therapy against FAP antigen expressing cancer, Ratio Therapeutics is recruiting for a Phase 1/2 Open-label Study (NCT07156565).
Eligibility requires history of relapse and refractory soft tissue sarcoma. Recruiting in US and Canada.
A key limiting factor for dietary use of single cell protein is the high mass fraction of nucleic acid, which limits daily consumption due to uric acid production during metabolism. High rates of RNA synthesis are unfortunately necessary for high protein productivity.
The paper notes:
>It is important to note that MP products often contain elevated levels of nucleic acids, constituting ~8% of the dry weight [17], which necessitates consideration when assessing their suitability for human consumption. To address this, a heat treatment process is employed at the end of fermentation that reduces the nucleic acid content in the fermented biomass to below 0.75/100 g, while simultaneously deactivating protease activity and F. venenatum biomass. However, this procedure has been observed to induce cell membrane leakage and a substantial loss of biomass, as evidenced in the Quorn production process [17], which also utilizes F. venenatum as the MP producer. Our experimental trials have encountered similar challenges, achieving a biomass yield of merely ~35%, and observed that heating process increased the relative protein and chitin content (Figure 2D,E), which may be related to the effect of membrane leakage, while the intracellular protein of the FCPD engineered strain was less likely to be lost to the extracellular. Thus, concentrating the fermentation broth to enhance protein and amino acids content in successive steps to produce a highly nutritious water-soluble fertilizer appears to be an effective strategy for adding value to the process (Figure 1).
The challenges of developing economic single cell protein products, that are suitable for human consumption, are described in chapter 3 here:
https://github.com/iOfficeAI/OfficeCLI
https://github.com/rcarmo/python-office-mcp-server
https://github.com/espressoplease/smalldocs
It would be great to see some benchmarks for token use, correctness, features etc.