Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

AI usage meter for users: per-run token/credit/cost popover and per-account counters (after billing is complete)

Open
  • Is something wrong?
#435 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
4/5
Estimated time
3-5 days
Newbie friendliness
25/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Domain
ai, api, backend, frontend, payments

Research direction

Start from ADR-008 and #173 (TokenUsage with cached tokens) to understand measured usage. The first piece is Krizaka UI's UsageMeter and UsageSummary components (stories, EN/FR, a11y) once krizaka-billing#72 ledger lands. Done means a popover shows per-run token/credit/cost breakdown and the account page shows period counters with CSV export, wired into all AI results.

Written by the indexing model from the issue text.

Description

public

Why (founder, 2026-10-30, screenshot of an AI studio's per-request usage popover)

« Issue pour ce type d'affiche pour les users qui utilisent des features d'IA : leurs compteurs. À ajouter dans le backlog, moins prioritaire, à faire quand le module bills sera complet. »

The reference shows, for one generation: input tokens (by modality), output tokens (by modality), thought tokens, the total against a limit (87 446 / 131 072), and an estimated cost per part and in total, labelled as an estimate.

What we want: a usage meter wherever a user runs an AI feature

In Orazaka's console and chat, Akinaza's AI photos and analytics (ADR-008), and Yomizaka's checks:

  • Per run: a small popover on the result, showing:
    • input, output, thought and cached tokens (by modality: text, image, video, audio);
    • the context used against the model's limit;
    • credits charged;
    • the estimated cost, labelled « estimation », in the user's currency.
  • Per account: the period's counters, credits left against the plan quota (or the admin-set quota), and a history with a CSV export.
  • Honest numbers: they come from the engine's real usage (Orazaka run usage and observations: TokenUsage with cached tokens, #173), never computed in the UI. The cost uses the price table the admin maintains (rule « admin configures everything »). Credits come from krizaka-billing.
  • Privacy: counts only, never prompt or answer text (same rule as the telemetry).

Pieces

  • Krizaka UI: UsageMeter (popover with the per-part breakdown, a limit bar and an estimate badge) and UsageSummary (period counters and a quota bar), with stories, both themes, EN/FR and a11y.
  • Orazaka: a per-run usage endpoint and a per-account period aggregate (already measured; expose them).
  • krizaka-billing: credits per run and the quota per plan, with the cached-read billing unit (krizaka-billing#72).
  • Products: wire the meter into each AI result and into the account page.

Order

Backlog, lower priority: after krizaka-billing is complete (#72 and the credits ledger), then Krizaka UI pieces → Orazaka endpoints → products.

Dominant language
No language data
Stars
0
Forks
0
Avg merge
2h 49m
Merged PRs (30d)
212

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from krizaka/.github

All issues in krizaka/.github

Similar issues

More AI Infra & Agents issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.