Local evidence dashboard
Compare content-free token metadata with observed quota movement, see what the model can and cannot explain, and improve the estimate over time.
Open TiboTattle from Applications and use the local dashboard it opens. Nothing has been presented as your real usage.
First-time setup
A normal website cannot read Codex files. The TiboTattle Mac app runs a private loopback companion, reads metadata locally, and opens your dashboard in a separate local tab.
This check only reconnects to the current page. If the app opened a different local dashboard tab, continue there.
You may close this hosted browser tab at any time. During an analysis, keep the TiboTattle app open; closing the app stops its local companion, while the last durable checkpoint remains available when you reopen it.
Local setup
TiboTattle checks only whether the expected Codex folders and its private state directory are available. It does not send or display prompts, responses, commands, paths, or account identifiers.
A useful headline often appears in seconds. The first deep pass can take a few minutes; later updates are normally faster. Work stops or checkpoints at a fixed bound, so a large Codex history may continue in another pass. You may close the browser tab; keep the Usage Monitor app open while analysis runs.
01 · Overview
Current quota observations and API-price-equivalent usage from your local evidence.
API-price equivalent
This applies public API prices to non-overlapping local token increments. It is not a subscription charge or a published dollar limit.
This activity total can exceed the inferred weekly limit: it spans a calendar period, while the weekly estimate describes one observed reset track and may cross resets, credits, or account changes.
Measured versus calculated
More observations are required before a useful comparison can be made.
This fit uses API prices as a measuring stick. It is not a provider-published dollar allowance.
02 · Timeline
Replay-safe local increments · browser-local time · current API prices
Analyze local usage to build recent content-free usage buckets.
Browser-local chart axis · exact local and UTC times are listed below
Connect the local companion or choose the labeled demo.
Residuals
Residual = observed quota change minus cost-implied change. Positive values mean the allowance fell faster than the token model predicted.
Inspect exact periods
| Exact local / UTC time | Observed | Expected | Difference | Evidence state |
|---|---|---|---|---|
| No periods loaded. | ||||
03 · Weekly allowance
Historical reset-series fits, plotted when each estimate became observable and with partial extrapolations kept visibly separate.
The estimate will appear when enough quota transitions can be matched to priced usage.
History
Points are independent reset-series fits, not a continuous balance. The shaded band is variation across reset series; per-row sensitivity remains in the evidence table. Scroll horizontally on a narrow screen.
Validation
There is not yet enough comparable weekly evidence to call a change.
The change statement follows the date and evidence filters above. Error metrics use all qualifying held-out observations when available.
Each row matches API-price-equivalent usage to a decrease in the provider-reported seven-day allowance, then scales that movement to an inferred full 100 percentage-point allowance. Partial rows are diagnostics and must not be read as completed weeks.
Week-by-week evidence
| Evidence available / reset due | Observed API equivalent | Quota decrease | Implied full allowance | Within-reset sensitivity | Usable observations | Coverage quality | Caveats |
|---|---|---|---|---|---|---|---|
| No weekly evidence loaded. | |||||||
04 · Cost accounting
Every local usage increment is repriced with the public Standard API rate card. Subscription Fast mode remains a separate observed attribute.
Token components
Output text excludes reasoning tokens. Combined output is used only when a source does not provide that split, and is never added again.
Models
| Model | Usage increments | Tokens | API equivalent |
|---|
Any unrecognized model is kept as an explicit overflow row. Its tokens remain visible, but no price is invented.
Subscription speed
Percentages are API-price-equivalent cost shares; counts are non-overlapping usage increments. Not recorded means an older rollout did not retain this setting and is not inferred as Standard.
Surface
Percentages are API-price-equivalent cost shares; counts are non-overlapping usage increments, not messages, turns, or every raw log row.
These fields audit the estimator; they are not provider billing categories and may be absent in current logs.
Lineage
Tool classes
API service tier
Reasoning effort
05 · Monitoring gaps
Honest coverage is part of the result. Blind spots, unknown models, missing prices, resets, and unobserved surfaces remain explicit.
Signal coverage
Each row is an independent signal. The badge is a count of signals, not an overall percentage score.
Needs investigation
These links open separately generated diagnostic HTML in a new tab. They stay on this computer and do not collect additional telemetry.
Help improve the estimate
Contribute content-free pseudonymous metadata so the community can measure how Codex limits differ over time. Prompts, responses, commands, paths, account names and credentials never leave this Mac.
Timestamps, token counts, model declarations or opaque model fingerprints, subscription speed, coarse tool and surface categories, and observed quota percentages.
No message content, reasoning text, filenames, URLs, commands, account names, email addresses, raw identifiers or credentials.
Sign in for hosted contribution Not signed in
Hosted participation requires signing in with Google or Apple. The service stores only an irreversible hash of that sign-in — never your email or name — and this page keeps the sign-in token in memory only. Local-only use needs no account.
Off until you explicitly consent. No daemon or login item is installed; checks run only while TiboTattle is open.
06 · Data & privacy
Raw logs are read by the loopback companion, not the webpage. A contribution must be a privacy-stripped export and is validated again by the server. Hosted participation requires signing in with Google or Apple; the service stores only an irreversible hash of that sign-in, never your email or name. Local-only use needs no account.
Browser validation is a preflight. The Worker decrypts into bounded memory, validates the same closed schema again, rejects content canaries, and recalculates API-price-equivalent values before D1 ingestion.
Local collector
The local companion has not started a bounded local analysis.
Raw log contents and source paths never enter this page.
You may close this browser tab while analysis continues. Keep the TiboTattle app open; closing it ends the local companion and preserves the last durable checkpoint for a later resume.
Review before the first send
Preparation reads the latest bounded interval locally and performs no upload. Inspect the concise summary, expand the exact JSON if wanted, then confirm the first send.
Prepare and review evidence
Analyze local usage to estimate the number of privacy-safe records and upload batches before preparation.
Nothing is sent until you separately inspect and choose a send action. A dense seven-day selection may exceed the single reviewed-set safety cap; if so, choose 24 hours instead.
Choose a .json file created by TiboTattle. Do not
choose a .jsonl file, a sessions folder, or a raw Codex log.
Contribution activity
The local queue contains metadata about committed, privacy-verified batches only. It never contains prompts, responses, source paths, account names, browser sessions, or device secrets.
Automatic checks run at most every 6 hours while TiboTattle is open and only after explicit consent. No login item, LaunchAgent, daemon or silent background process is installed.
Optional community service
This optional service is separate from the local collector above. Live readiness verifies database and encrypted-object access, fresh retention and restore replay, object reconciliation, and aggregate rebuild state. The repeatable backend suite covers the complete ingestion, isolation, export, and deletion lifecycle.
Your contributed evidence
Accepted intervals and private calculations appear here. Personal reporting remains in this local app; contribution helps improve the delayed community view.
No accepted contribution is associated with this browser yet.
Permanently remove the content-free pseudonymous metadata associated with this hosted contribution session.
Delayed community evidence
Checking for the latest delayed weekly snapshot…
Community values use a fixed delay, independent per-cell support, participant-level clipping, and coarse rounding. A sealed revision is never rewritten; privacy-affecting deletion withdraws it and schedules a replacement revision without the deleted source.