chore(release): 0.39.43

[pitimon]

Published to npm as @ipv9/tokentracker-cli@0.39.43 before opening this — the publish job self-skips a version that already exists, so CI stays green.

Ships

#98 (issue #97) — GitHub Copilot quota now reports the premium-request count (158/300, or 142/300 in Remaining mode) instead of a percentage the reader has to convert. The numbers were already being fetched from api.github.com and discarded after one division.

The QA gate earned its keep

An independent Codex pass at xhigh returned SHIP: NO on the first cut. All three findings were real and all three were mine:

1. The bar and its caption could be drawn from different fields.percent_remaining: 30 with entitlement: 300, remaining: 72 produced a 70%-wide bar captioned 228/300 — which is 76%. The percentage is now derived from the counts whenever both arrive.
2. Number(null) is 0, so a missing remaining reported the entire allowance consumed: 300/300 beside a percentage saying 10% used. Counts now require a real finite number.
3. The dashboard clamped one end only, so an unclamped payload rendered 312/300 in Used mode from used: 312and in Remaining mode from used: -12. Clamping into [0, limit] now happens before the mode flip — which is what my own discarded comment had claimed to handle.

Each is closed with a regression test built from the exact reproduction. A narrow re-check returned SHIP: YES, plus two non-blocking stale comments, also fixed.

First real use of #96's lockstep hook

npm version 0.39.43 synced TokenTrackerBar/project.yml and TokenTrackerWin.csproj automatically. That drift is precisely what made 0.39.39–0.39.42 unable to cut a desktop build.

This PR also proves the #96 branch protection

I attempted the direct push to main first, as this repo has always done for release commits. It was rejected:

remote: error: GH006: Protected branch update failed for refs/heads/main.
remote: - Required status check "ci:local" is expected.

That closes the one verification gap #96 shipped with — it was config-read-back only, because I would not land an unreviewed commit on main to test it. Release commits now go through a PR like everything else.

Test plan

• npm run ci:local — exit 0, 841/841 root, 255 dashboard
• Pricing release gate on the 0.39.43 build: unpriced_models: []
• npm pack --dry-run contains dashboard/dist/index.html + both service scripts
• npm view confirms 0.39.43 on the registry
• Seed re-vendored by prepublishOnly: 2,522 models, 0 rate changes, only _meta.generated_at
• Copilot counts still not verified against a live account — this machine has no token (configured: false)

[pitimon]

Published to npm as @ipv9/tokentracker-cli@0.39.43 before opening this — the publish job self-skips a version that already exists, so CI stays green.

Ships

#98 (issue #97) — GitHub Copilot quota now reports the premium-request count (158/300, or 142/300 in Remaining mode) instead of a percentage the reader has to convert. The numbers were already being fetched from api.github.com and discarded after one division.

The QA gate earned its keep

An independent Codex pass at xhigh returned SHIP: NO on the first cut. All three findings were real and all three were mine:

1. The bar and its caption could be drawn from different fields.percent_remaining: 30 with entitlement: 300, remaining: 72 produced a 70%-wide bar captioned 228/300 — which is 76%. The percentage is now derived from the counts whenever both arrive.
2. Number(null) is 0, so a missing remaining reported the entire allowance consumed: 300/300 beside a percentage saying 10% used. Counts now require a real finite number.
3. The dashboard clamped one end only, so an unclamped payload rendered 312/300 in Used mode from used: 312and in Remaining mode from used: -12. Clamping into [0, limit] now happens before the mode flip — which is what my own discarded comment had claimed to handle.

Each is closed with a regression test built from the exact reproduction. A narrow re-check returned SHIP: YES, plus two non-blocking stale comments, also fixed.

First real use of #96's lockstep hook

npm version 0.39.43 synced TokenTrackerBar/project.yml and TokenTrackerWin.csproj automatically. That drift is precisely what made 0.39.39–0.39.42 unable to cut a desktop build.

This PR also proves the #96 branch protection

I attempted the direct push to main first, as this repo has always done for release commits. It was rejected:

remote: error: GH006: Protected branch update failed for refs/heads/main.
remote: - Required status check "ci:local" is expected.

That closes the one verification gap #96 shipped with — it was config-read-back only, because I would not land an unreviewed commit on main to test it. Release commits now go through a PR like everything else.

Test plan

• npm run ci:local — exit 0, 841/841 root, 255 dashboard
• Pricing release gate on the 0.39.43 build: unpriced_models: []
• npm pack --dry-run contains dashboard/dist/index.html + both service scripts
• npm view confirms 0.39.43 on the registry
• Seed re-vendored by prepublishOnly: 2,522 models, 0 rate changes, only _meta.generated_at
• Copilot counts still not verified against a live account — this machine has no token (configured: false)

[pitimon]

Published to npm as @ipv9/tokentracker-cli@0.39.43 before opening this — the publish job self-skips a version that already exists, so CI stays green.

Ships

#98 (issue #97) — GitHub Copilot quota now reports the premium-request count (158/300, or 142/300 in Remaining mode) instead of a percentage the reader has to convert. The numbers were already being fetched from api.github.com and discarded after one division.

The QA gate earned its keep

An independent Codex pass at xhigh returned SHIP: NO on the first cut. All three findings were real and all three were mine:

1. The bar and its caption could be drawn from different fields.percent_remaining: 30 with entitlement: 300, remaining: 72 produced a 70%-wide bar captioned 228/300 — which is 76%. The percentage is now derived from the counts whenever both arrive.
2. Number(null) is 0, so a missing remaining reported the entire allowance consumed: 300/300 beside a percentage saying 10% used. Counts now require a real finite number.
3. The dashboard clamped one end only, so an unclamped payload rendered 312/300 in Used mode from used: 312and in Remaining mode from used: -12. Clamping into [0, limit] now happens before the mode flip — which is what my own discarded comment had claimed to handle.

Each is closed with a regression test built from the exact reproduction. A narrow re-check returned SHIP: YES, plus two non-blocking stale comments, also fixed.

First real use of #96's lockstep hook

npm version 0.39.43 synced TokenTrackerBar/project.yml and TokenTrackerWin.csproj automatically. That drift is precisely what made 0.39.39–0.39.42 unable to cut a desktop build.

This PR also proves the #96 branch protection

I attempted the direct push to main first, as this repo has always done for release commits. It was rejected:

remote: error: GH006: Protected branch update failed for refs/heads/main.
remote: - Required status check "ci:local" is expected.

That closes the one verification gap #96 shipped with — it was config-read-back only, because I would not land an unreviewed commit on main to test it. Release commits now go through a PR like everything else.

Test plan

• npm run ci:local — exit 0, 841/841 root, 255 dashboard
• Pricing release gate on the 0.39.43 build: unpriced_models: []
• npm pack --dry-run contains dashboard/dist/index.html + both service scripts
• npm view confirms 0.39.43 on the registry
• Seed re-vendored by prepublishOnly: 2,522 models, 0 rate changes, only _meta.generated_at
• Copilot counts still not verified against a live account — this machine has no token (configured: false)

[pitimon]

Published to npm as @ipv9/tokentracker-cli@0.39.43 before opening this — the publish job self-skips a version that already exists, so CI stays green.

Ships

#98 (issue #97) — GitHub Copilot quota now reports the premium-request count (158/300, or 142/300 in Remaining mode) instead of a percentage the reader has to convert. The numbers were already being fetched from api.github.com and discarded after one division.

The QA gate earned its keep

An independent Codex pass at xhigh returned SHIP: NO on the first cut. All three findings were real and all three were mine:

1. The bar and its caption could be drawn from different fields.percent_remaining: 30 with entitlement: 300, remaining: 72 produced a 70%-wide bar captioned 228/300 — which is 76%. The percentage is now derived from the counts whenever both arrive.
2. Number(null) is 0, so a missing remaining reported the entire allowance consumed: 300/300 beside a percentage saying 10% used. Counts now require a real finite number.
3. The dashboard clamped one end only, so an unclamped payload rendered 312/300 in Used mode from used: 312and in Remaining mode from used: -12. Clamping into [0, limit] now happens before the mode flip — which is what my own discarded comment had claimed to handle.

Each is closed with a regression test built from the exact reproduction. A narrow re-check returned SHIP: YES, plus two non-blocking stale comments, also fixed.

First real use of #96's lockstep hook

npm version 0.39.43 synced TokenTrackerBar/project.yml and TokenTrackerWin.csproj automatically. That drift is precisely what made 0.39.39–0.39.42 unable to cut a desktop build.

This PR also proves the #96 branch protection

I attempted the direct push to main first, as this repo has always done for release commits. It was rejected:

remote: error: GH006: Protected branch update failed for refs/heads/main.
remote: - Required status check "ci:local" is expected.

That closes the one verification gap #96 shipped with — it was config-read-back only, because I would not land an unreviewed commit on main to test it. Release commits now go through a PR like everything else.

Test plan

• npm run ci:local — exit 0, 841/841 root, 255 dashboard
• Pricing release gate on the 0.39.43 build: unpriced_models: []
• npm pack --dry-run contains dashboard/dist/index.html + both service scripts
• npm view confirms 0.39.43 on the registry
• Seed re-vendored by prepublishOnly: 2,522 models, 0 rate changes, only _meta.generated_at
• Copilot counts still not verified against a live account — this machine has no token (configured: false)

[pitimon]

Published to npm as @ipv9/tokentracker-cli@0.39.43 before opening this — the publish job self-skips a version that already exists, so CI stays green.

Ships

#98 (issue #97) — GitHub Copilot quota now reports the premium-request count (158/300, or 142/300 in Remaining mode) instead of a percentage the reader has to convert. The numbers were already being fetched from api.github.com and discarded after one division.

The QA gate earned its keep

An independent Codex pass at xhigh returned SHIP: NO on the first cut. All three findings were real and all three were mine:

1. The bar and its caption could be drawn from different fields.percent_remaining: 30 with entitlement: 300, remaining: 72 produced a 70%-wide bar captioned 228/300 — which is 76%. The percentage is now derived from the counts whenever both arrive.
2. Number(null) is 0, so a missing remaining reported the entire allowance consumed: 300/300 beside a percentage saying 10% used. Counts now require a real finite number.
3. The dashboard clamped one end only, so an unclamped payload rendered 312/300 in Used mode from used: 312and in Remaining mode from used: -12. Clamping into [0, limit] now happens before the mode flip — which is what my own discarded comment had claimed to handle.

Each is closed with a regression test built from the exact reproduction. A narrow re-check returned SHIP: YES, plus two non-blocking stale comments, also fixed.

First real use of #96's lockstep hook

npm version 0.39.43 synced TokenTrackerBar/project.yml and TokenTrackerWin.csproj automatically. That drift is precisely what made 0.39.39–0.39.42 unable to cut a desktop build.

This PR also proves the #96 branch protection

I attempted the direct push to main first, as this repo has always done for release commits. It was rejected:

remote: error: GH006: Protected branch update failed for refs/heads/main.
remote: - Required status check "ci:local" is expected.

That closes the one verification gap #96 shipped with — it was config-read-back only, because I would not land an unreviewed commit on main to test it. Release commits now go through a PR like everything else.

Test plan

• npm run ci:local — exit 0, 841/841 root, 255 dashboard
• Pricing release gate on the 0.39.43 build: unpriced_models: []
• npm pack --dry-run contains dashboard/dist/index.html + both service scripts
• npm view confirms 0.39.43 on the registry
• Seed re-vendored by prepublishOnly: 2,522 models, 0 rate changes, only _meta.generated_at
• Copilot counts still not verified against a live account — this machine has no token (configured: false)

[pitimon]

Published to npm as @ipv9/tokentracker-cli@0.39.43 before opening this — the publish job self-skips a version that already exists, so CI stays green.

Ships

#98 (issue #97) — GitHub Copilot quota now reports the premium-request count (158/300, or 142/300 in Remaining mode) instead of a percentage the reader has to convert. The numbers were already being fetched from api.github.com and discarded after one division.

The QA gate earned its keep

An independent Codex pass at xhigh returned SHIP: NO on the first cut. All three findings were real and all three were mine:

1. The bar and its caption could be drawn from different fields.percent_remaining: 30 with entitlement: 300, remaining: 72 produced a 70%-wide bar captioned 228/300 — which is 76%. The percentage is now derived from the counts whenever both arrive.
2. Number(null) is 0, so a missing remaining reported the entire allowance consumed: 300/300 beside a percentage saying 10% used. Counts now require a real finite number.
3. The dashboard clamped one end only, so an unclamped payload rendered 312/300 in Used mode from used: 312and in Remaining mode from used: -12. Clamping into [0, limit] now happens before the mode flip — which is what my own discarded comment had claimed to handle.

Each is closed with a regression test built from the exact reproduction. A narrow re-check returned SHIP: YES, plus two non-blocking stale comments, also fixed.

First real use of #96's lockstep hook

npm version 0.39.43 synced TokenTrackerBar/project.yml and TokenTrackerWin.csproj automatically. That drift is precisely what made 0.39.39–0.39.42 unable to cut a desktop build.

This PR also proves the #96 branch protection

I attempted the direct push to main first, as this repo has always done for release commits. It was rejected:

remote: error: GH006: Protected branch update failed for refs/heads/main.
remote: - Required status check "ci:local" is expected.

That closes the one verification gap #96 shipped with — it was config-read-back only, because I would not land an unreviewed commit on main to test it. Release commits now go through a PR like everything else.

Test plan

• npm run ci:local — exit 0, 841/841 root, 255 dashboard
• Pricing release gate on the 0.39.43 build: unpriced_models: []
• npm pack --dry-run contains dashboard/dist/index.html + both service scripts
• npm view confirms 0.39.43 on the registry
• Seed re-vendored by prepublishOnly: 2,522 models, 0 rate changes, only _meta.generated_at
• Copilot counts still not verified against a live account — this machine has no token (configured: false)

[pitimon]

Published to npm as @ipv9/tokentracker-cli@0.39.43 before opening this — the publish job self-skips a version that already exists, so CI stays green.

Ships

#98 (issue #97) — GitHub Copilot quota now reports the premium-request count (158/300, or 142/300 in Remaining mode) instead of a percentage the reader has to convert. The numbers were already being fetched from api.github.com and discarded after one division.

The QA gate earned its keep

An independent Codex pass at xhigh returned SHIP: NO on the first cut. All three findings were real and all three were mine:

1. The bar and its caption could be drawn from different fields.percent_remaining: 30 with entitlement: 300, remaining: 72 produced a 70%-wide bar captioned 228/300 — which is 76%. The percentage is now derived from the counts whenever both arrive.
2. Number(null) is 0, so a missing remaining reported the entire allowance consumed: 300/300 beside a percentage saying 10% used. Counts now require a real finite number.
3. The dashboard clamped one end only, so an unclamped payload rendered 312/300 in Used mode from used: 312and in Remaining mode from used: -12. Clamping into [0, limit] now happens before the mode flip — which is what my own discarded comment had claimed to handle.

Each is closed with a regression test built from the exact reproduction. A narrow re-check returned SHIP: YES, plus two non-blocking stale comments, also fixed.

First real use of #96's lockstep hook

npm version 0.39.43 synced TokenTrackerBar/project.yml and TokenTrackerWin.csproj automatically. That drift is precisely what made 0.39.39–0.39.42 unable to cut a desktop build.

This PR also proves the #96 branch protection

I attempted the direct push to main first, as this repo has always done for release commits. It was rejected:

remote: error: GH006: Protected branch update failed for refs/heads/main.
remote: - Required status check "ci:local" is expected.

That closes the one verification gap #96 shipped with — it was config-read-back only, because I would not land an unreviewed commit on main to test it. Release commits now go through a PR like everything else.

Test plan

• npm run ci:local — exit 0, 841/841 root, 255 dashboard
• Pricing release gate on the 0.39.43 build: unpriced_models: []
• npm pack --dry-run contains dashboard/dist/index.html + both service scripts
• npm view confirms 0.39.43 on the registry
• Seed re-vendored by prepublishOnly: 2,522 models, 0 rate changes, only _meta.generated_at
• Copilot counts still not verified against a live account — this machine has no token (configured: false)

[pitimon]

Published to npm as @ipv9/tokentracker-cli@0.39.43 before opening this — the publish job self-skips a version that already exists, so CI stays green.

Ships

#98 (issue #97) — GitHub Copilot quota now reports the premium-request count (158/300, or 142/300 in Remaining mode) instead of a percentage the reader has to convert. The numbers were already being fetched from api.github.com and discarded after one division.

The QA gate earned its keep

An independent Codex pass at xhigh returned SHIP: NO on the first cut. All three findings were real and all three were mine:

1. The bar and its caption could be drawn from different fields.percent_remaining: 30 with entitlement: 300, remaining: 72 produced a 70%-wide bar captioned 228/300 — which is 76%. The percentage is now derived from the counts whenever both arrive.
2. Number(null) is 0, so a missing remaining reported the entire allowance consumed: 300/300 beside a percentage saying 10% used. Counts now require a real finite number.
3. The dashboard clamped one end only, so an unclamped payload rendered 312/300 in Used mode from used: 312and in Remaining mode from used: -12. Clamping into [0, limit] now happens before the mode flip — which is what my own discarded comment had claimed to handle.

Each is closed with a regression test built from the exact reproduction. A narrow re-check returned SHIP: YES, plus two non-blocking stale comments, also fixed.

First real use of #96's lockstep hook

npm version 0.39.43 synced TokenTrackerBar/project.yml and TokenTrackerWin.csproj automatically. That drift is precisely what made 0.39.39–0.39.42 unable to cut a desktop build.

This PR also proves the #96 branch protection

I attempted the direct push to main first, as this repo has always done for release commits. It was rejected:

remote: error: GH006: Protected branch update failed for refs/heads/main.
remote: - Required status check "ci:local" is expected.

That closes the one verification gap #96 shipped with — it was config-read-back only, because I would not land an unreviewed commit on main to test it. Release commits now go through a PR like everything else.

Test plan

• npm run ci:local — exit 0, 841/841 root, 255 dashboard
• Pricing release gate on the 0.39.43 build: unpriced_models: []
• npm pack --dry-run contains dashboard/dist/index.html + both service scripts
• npm view confirms 0.39.43 on the registry
• Seed re-vendored by prepublishOnly: 2,522 models, 0 rate changes, only _meta.generated_at
• Copilot counts still not verified against a live account — this machine has no token (configured: false)