c5huracan

4 ms·

1. 1.▲
   by c5huracan 7mo ago

   The bottleneck isn't "how do I define my agent." It's "how do agents find the right tool for their task." I run a search service that 110+ agents use. They don't browse catalogs or read specs. They describ
2. 2.▲
   by c5huracan 7mo ago

   The "no meaningful benchmark for good agentic session performance" point resonates. Success varies so much by task type that a single metric is almost meaningless. A 60-second documentation lookup and a 30-minute refactoring sessi
3. 3.▲
   by c5huracan 7mo ago

   Use Vault and use a proxy. They address different problems. Vault protects keys at rest, but the agent still gets them at runtime. The proxy keeps the key away from the agent entirely, which closes key leakage. But a prompt-injected agent c
4. 4.▲
   by c5huracan 7mo ago

   The trust scoring layer is the interesting part here. The agent ecosystem has a discovery problem and a trust problem, and most tools today only tackle discovery. Being able to evaluate reliability before you connect changes the calculus. C
5. 5.▲
   Show HN: Search 7,500 MCP servers across NPM, PyPI, and the official registry(api.rhdxm.com)

   3 points by c5huracan 7mo ago|0 comments