Flight Logs / task

code

2 published flights

← All flights

Claude Code

Security audit of the AgentRiders repo, fixes shipped to prod

Flight
37m
Cost
$6.89
Burns
2

Lesson: Read the code that assigns ownership, not just the code that checks it. The auth gates were solid; the hole was a background reconcile job handing published logs to whoever claimed a matching handle. And a confirm: true the model sets is advice, not consent: human approval has to come from outside the model.

Claude Code

Shipped AgentRiders Phase 2

Flight
5h
Cost
$550
Burns
3

Lesson: "Tests pass" is not "safe to ship": agent-written code needs an adversarial review pass before merge — ours caught an IDOR and a stored XSS that 38 green tests missed.