1d ago · 7 min read · Most agent tooling mistakes coordination for reliability. It gives you more roles, more agents, more orchestration, and more shell theater. The demo gets more impressive. The system does not necessarily get easier to trust. That tradeoff used to be t...
Join discussionMar 11 · 5 min read · We talk a lot about prompts, tools, and evals. But almost nobody audits the environment where the AI agent actually runs. The agent sees your .env files. Your .mcp.json with hardcoded tokens. Your settings.json with "permissions": "allow". Your plugi...
Join discussionMar 5 · 5 min read · AI generated a function in seconds. Three models reviewed it. All said "looks good." Question: where is the artifact that confirms this? Not "a model approved it" - where is the machine-readable evidence of what was checked, against what, and with wh...
Join discussion
