Tarcroi

5 ms·

1. 1.▲
   by Tarcroi 5mo ago

   I find it improbable that a car could have gotten through in a marathon as big as London's, with its incredible organization. In any case, having run quite a few marathons, I can tell that a marathon is anything but a distraction, it&#
2. 2.▲
   by Tarcroi 5mo ago

   This coincides with Anthropic's peak-hour announcement (March 26th). Could the throttling be partly a response to infrastructure load that was itself inflated by the TTL regression?
3. 3.▲
   by Tarcroi 5mo ago

   Different thing. SkillKit distributes skills to agents. Skrun runs them as APIs.
4. 4.▲
   by Tarcroi 5mo ago

   As I mentioned in another comment here, I've been working on an open-source alternative. Multi-model, 5 providers with fallback. Happy to share the repo if you're interested.
5. 5.▲
   by Tarcroi 5mo ago

   I agree on the commodity point, that's why I went multi-model from start. The registry question is the one I'm thinking about the most. Right now it's flat. I plan to integrate usage data (success rates, cost, trust scores).
6. 6.▲
   by Tarcroi 5mo ago

   I've been building exactly this. it's Open source, multi-model (5 providers with fallback), from now, it runs locally but the architecture is designed for self-hosted deployment.
7. 7.▲
   by Tarcroi 5mo ago

   Ha, thanks for the correction! I'll remember that!
8. 8.▲
   by Tarcroi 5mo ago

   You're right. For now, it's only local. For a public deployment, the idea is to have sandboxes and verification steps. That won't completely eliminate the risk of prompt injection, but so far no solution has managed to comple
9. 9.▲
   by Tarcroi 5mo ago

   Both. In my opinion, an agent has a life cycle and needs observability.
10. 10.▲
    by Tarcroi 5mo ago

    Thanks! Sandbox deployment is planned in the roadmap. I already have a RuntimeAdapter interface in my architecture that I'll use to isolate the VMs. I'm doing exactly the same thing: I'm cross-referencing the models to challe
11. 11.▲
    by Tarcroi 5mo ago

    Can this be run a second time and compared against a previous audit?
12. 12.▲
    by Tarcroi 5mo ago

    Hi, I'm the "colleague", Impatient to have your feedback!