Skip to content

chore(release): prepare 0.8.2 - #212

Merged
pierreraby merged 1 commit into
mainfrom
release/0.8.2
Nov 11, 2026
Merged

pierreraby merged 1 commit into
mainfrom
release/0.8.2

Conversation

@pierreraby

Copy link
Copy Markdown
Collaborator

Release 0.8.2

Patch release for the reviewed November pricing merged in #211. Channel: stable (v0.8.2 → npm latest).

Why

/provider/v1/models advertised 97 models on 2026-11-11 (was 96). Three had no reviewed price, so 0.8.1 showed a zero estimate plus an unknown-price warning for every request using them, and the daily catalog workflow failed its pricing gate twice (runs 42838424841, 43037482750), reopening the maintenance incident #207.

What ships

  • claude-fable-5-2 at $8.00 / $40.00 / $0.20 / $10.00 per million tokens (input / output / cache read / cache write).
  • google/gemini-4-carbon at $2.00 / $10.00 / $0.10, no cache-write charge.
  • meta/muse-ember at $0.10 / $0.50 / $0.004, no cache-write charge.
  • Removed the zero prices for stealth/crouton-alpha and reflection/beam:free, no longer advertised.
  • Snapshots refreshed to 97 models, PRICING_LAST_VERIFIED = 2026-11-11.

No endpoint-routing, entitlement, credential or host-contract change. Static capability metadata remains command-code@1.88.5, so the three new models stay text-only without reasoning or effort flags until upstream ships them — they were already usable through the live API before this release.

Preflight (recorded before tagging)

  • npm dist-tags: latest=0.8.1, next=0.7.3-next.0 (unchanged baseline).
  • Latest stable GitHub Release: v0.8.1.
  • npm view pi-commandcode-provider@0.8.2 → E404 (unused version); no v0.8.2 tag locally or on the remote; no active release.yml run.

Local verification

  • npm test with PI_LOCAL_REQUIRED=1 OMP_COMPAT_REQUIRED=1 → PASS on real hosts (Pi 1.5.2, Oh My Pi 18.18.3, bun 1.4.4), all three protocol contracts on both hosts plus endpoint routing.
  • npm run check:commandcode-pricing → PASS, 97 models, no drift; npm run check:commandcode-catalog → metadata current at command-code@1.88.5.
  • npm run format:check, npm audit --audit-level=moderate (0 vulnerabilities), npm pack --dry-run (31 files), git diff --check. actionlint is not installed locally; the CI workflow job covers workflow YAML.
  • Post-merge proof of the fix: workflow_dispatch of the catalog workflow on 5542515 succeeded (run 43102876321) and the maintenance alert bot closed Maintenance: repeated catalog sync failures #207.

Mock APIs only: this does not claim live Command Code billing or entitlement verification.

Ship the reviewed November pricing from #211 to npm users. The Provider
API advertised three unpriced models (claude-fable-5-2,
google/gemini-4-carbon, meta/muse-ember), so 0.8.1 reported a zero
estimate and an unknown-price warning for them and the daily catalog
workflow failed its pricing gate, reopening #207.

The changelog section records the new rates, the removed
stealth/crouton-alpha and reflection/beam:free entries, and the fact
that static capability metadata stays at command-code@1.88.5.
@github-actions

Copy link
Copy Markdown
Contributor

Dependency Review

✅ No vulnerabilities or license issues or OpenSSF Scorecard issues found.

Scanned Files

None

@github-actions

Copy link
Copy Markdown
Contributor

Memory benchmark

Compared base 5542515 with PR head 20eac51 on the same GitHub-hosted darwin runner. Lower values are better.

Metric Base PR PR − Base Change
Stable RSS 107.3 MiB ± 1.0 MiB 106.8 MiB ± 0.6 MiB -0.1 MiB ± 1.5 MiB -0.1%
Physical footprint 60.7 MiB ± 1.4 MiB 60.5 MiB ± 0.8 MiB -0.3 MiB ± 1.6 MiB -0.5%
Physical peak 72.5 MiB ± 2.1 MiB 71.6 MiB ± 0.6 MiB +0.4 MiB ± 3.0 MiB +0.5%

Interpretation: No metric shows a clear Base-to-PR difference beyond its measured run-to-run variation.

Extension overhead above pi baseline

Metric Base overhead PR overhead Difference
Stable RSS +5.2 MiB ± 0.2 MiB +5.2 MiB ± 0.7 MiB -0.1 MiB ± 1.5 MiB
Physical footprint +2.5 MiB ± 1.2 MiB +2.7 MiB ± 0.7 MiB -0.3 MiB ± 1.6 MiB
Physical peak +2.8 MiB ± 2.0 MiB +3.1 MiB ± 0.9 MiB +0.4 MiB ± 3.0 MiB

Values are medians of 6 alternating, paired runs. The value after ± is the median absolute deviation (MAD). Each process was sampled 12 times after a 2500 ms warm-up.

Environment: pi 0.84.4, Bun 1.4.4, darwin arm64. RSS comes from ps; physical footprint and peak come from macOS footprint.

This is a comparative signal, not a pass/fail threshold. GitHub-hosted runner noise can affect absolute values.

@pierreraby
pierreraby merged commit 7616ecd into main Nov 11, 2026
15 checks passed
@pierreraby
pierreraby deleted the release/0.8.2 branch November 11, 2026 21:32
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Maintenance: repeated catalog sync failures

1 participant