A Ready-to-Use Prompt That Makes Your Agent Audit Its Own API Bills
gethackteam · x · 2026-09-24
gethackteam shares a practical agent prompt: compare two weeks of traces and invoices, find which tasks got pricier, separate price changes from usage changes (calls, retries, output length), and show the evidence — producing a cost report you can actually act on.
More from coding & agent
- Hot take: a 22-year-old fluent in coding agents beats a lazy senior dev — but watch out for 'slop grenades' — jobergum · 2026-09-24
- Blender MCP hits 29k stars as Opus 5.5 builds cities on medium effort — sidahuj · 2026-09-24
- AI Agent Swarm Reverse-Engineers 2001 GBA Game Snood Byte-for-Byte in Two Weeks — Aizkmusic · 2026-09-24
- AI sped up game development, but shipping a game people want to play is still a grind — AIandDesign · 2026-09-24
- Texting an agent to fix streetlights: Instinct automates civic ticket filing end to end — vaibhavbetter · 2026-09-24
- Cursor launches Rollouts: AI writes monitoring plans and verifies deployments before users see regressions — belce_dogru · 2026-09-24