docs: sync every doc surface with the v1.63 release

/document-release audit (4-lane, all claims verified against branch code):

- README: gstack-egress + gstack-context-bill rows in the standalone-binaries
  table; Privacy & Telemetry gains the receipted-egress bullet (attempted-
  egress framing per the shipped threat model).
- ARCHITECTURE: /health is liveness-only, POST /extension-token endpoint row
  + bootstrap mechanics paragraph; new Egress receipt ledger subsection under
  Security model; eval persistence covers the sharded runner, GSTACK_EVAL_DIR,
  and the finalized-run baseline rule.
- CLAUDE.md: sharded test scripts in Commands; sharded semantics in the
  detached-evals section; PTY skill seeding in the hermetic section; egress
  invariant block beside the other server-egress invariants; catalog-budget
  ceiling beside the 160KB token ceiling; project-tree entries for
  lib/egress-receipt.ts, lib/context-bill.ts, scripts/test-paid-shards.ts.
- CONTRIBUTING: seedSkills + live-tree seeding in the hermetic paragraph;
  sharded runner in detached runs; catalog-budget in the Tier 1 list.
- BROWSER: extension token bootstrap section, tunnel egress receipts section,
  identity-pin migration note in manual install.
- REMOTE_BROWSER_ACCESS: tunnel-start receipt bullet in the security model.
- gbrain docs: /sync-gbrain + brain-sync egress-receipt behavior documented;
  dead consumer-token instructions removed (consumer machinery deleted this
  release); new fail-closed refusal added to the error catalog.
- CHANGELOG: measured-vs-ceiling catalog numbers, contributor notes for the
  external-service tier move and the PTY single-line AskUserQuestion parser,
  release date.
- TODOS: /health token-distribution TODO resolved by this release, removed;
  port-wave follow-up sections re-labeled to the shipped version.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This commit is contained in:
Garry Tan
2026-08-13 18:28:36 -07:00
co-authored by Claude Fable 5
parent 562273d793
commit e7110aac93
11 changed files with 169 additions and 54 deletions
+12 -2
View File
@@ -1,6 +1,6 @@
# Changelog
## [1.63.0.0] - 2026-08-12
## [1.63.0.0] - 2026-08-13
**Everything gstack sends off your machine now leaves a receipt you can read.**
**And the eval harness stopped grading itself a passing grade.**
@@ -25,7 +25,7 @@ Source: the assembled branch (`git log 1.62.0.0..HEAD`), the free suite
| gstack-owned off-machine sinks with a receipt | 0 | every enumerated sink | tripwire-enforced, zero exceptions |
| Eval "no regressions" lines that were self-comparisons | every one | 0 | the harness compared runs against their own in-progress accumulator |
| Paid gate runner isolation | one process, one hung file kills the tier | one process per file, group-SIGKILL on stall | + never-started accounting |
| Discovery catalog budget | unenforced | pinned at 1,105 token-equivalents | ratchet-protocol on every skill add |
| Discovery catalog budget | unenforced | 1,105 token-equivalents measured, 1,150 ceiling | ratchet-protocol on every skill add |
| Browser `/health` endpoint | served the root auth token to any localhost caller in headed mode | serves no token in any mode | token bootstrap moved to a pinned-origin POST |
The eval-store line is the one that matters most for anyone hacking on gstack:
@@ -95,6 +95,16 @@ bug fix and the port shortlist were selected and hardened for upstream.
the seeder, context-bill, and the catalog gate.
- `CLAUDE.md`'s compiled-binaries note corrected: the `browse/dist` binaries have
been untracked since v0.11.16.0, so they no longer appear in `git status`.
- External-service E2E tests (Codex, Gemini, benchmark providers) are declared
periodic-tier with the canonical whole-file guard, so the merge-blocking gate
never waits on a third-party CLI. The Codex runner passes
`--skip-git-repo-check` (now required in non-git working dirs) and the Gemini
runner classifies an unusable CLI (removed flags, retired auth paths) as a
skip instead of a false failure.
- The PTY test runner parses AskUserQuestion prompts that reflow onto a single
logical line and strips DEC cursor-visibility residue, pinned by
`test/pty-askuserquestion-single-line.test.ts` — the failure class that
previously ate a gate test's whole time budget.
- New follow-ups filed in `TODOS.md`: egress ledger rotation (chain-genesis
records), a launch-nonce token bootstrap, and eval-watch shard-awareness.