Gate paid-API spend on relevance and budget

01 · Role in the system

What Jev does here

The simplest gate asks Jev whether the query is actually relevant before any money moves - on-topic scored 0.96 and the paid call ran for a cent; off-topic scored 0.01 and nothing was spent. A typed choice picks exactly one route from a fixed set of three paid endpoints and cannot name a fourth. A market-pulse agent branches a question across three sources with parallel boolean checks, calls only the needed ones, and a second Jev pass verifies the fetched data covers what was asked. The enterprise pattern layers a second independent gate: three simulated teams share one key with per-team budget caps, and the verified run shows a relevant query blocked for exceeding Growth's cap while an irrelevant one was separately blocked by relevance - proving the two gates decide independently - while other teams' spend continued unaffected.

02 · Control boundary

Where Jev sits

Jev as the pre-spend gate: relevance Noul below threshold skips the call, choice routes among fixed endpoints, verify pass checks coverage, and budget checks run independently of relevance.

Code owns the loop, permissions, thresholds, validation, and side effects. Jev owns only the bounded judgments described above.

03 · Known limits

What this evidence does not prove

• All runs are author-verified POCs with simulated teams; no production traffic has passed through the gates.
• The 0.7 relevance threshold and team caps are the author's policy choices, not tuned benchmarks.
• Calls run through the Vercel AI Gateway rather than the native endpoint.

04 · Attribution

Public sources

This is a Community record: the project was published by a third-party community author.

• sprah ↗Community · github · public · checked 2026-09-27