The fast.md log_to_state fix changes an installed workflow file, so the per-
runtime golden-install-parity fixtures (18 runtimes) and the workflow-size
baseline are recaptured. Diff is exactly one entry per fixture (the fast.md
hash) and one baseline byte count — no spurious drift.
The first draft banned the literal /NF-1/ anywhere in the block, but the
explanatory comment legitimately references 'NF-1' to document the off-by-one
root cause. Match the COL_COUNT awk assignment specifically so the structural
guard targets the executable formula, not the prose.
Replaces the removed prose-regex test (bug-3805-*) that let the off-by-one
ship green. Extracts the actual bash block deployed in fast.md's log_to_state
step and EXECUTES it against real STATE.md fixtures, asserting on the
filesystem result: a row is appended with a cell count matching the header for
both the 5-column (non-validate) and 6-column (validate) schemas, and an
unrecognized schema still skips with a warning.
Four require() examples in commands/gsd/surface.md used bare
'gsd-core/...' specifiers that Node cannot resolve (wrong package
name + runtime-mirror layout off module path). Now derives the path
from runtimeConfigDir. Also fixes reinstall hint from
'npm i -g gsd-core' to 'npm i -g @opengsd/gsd-core'.
scanEntropyAnomalies + shannonEntropy were dead exports with zero
production callers — the live hooks (gsd-prompt-guard.js,
gsd-read-injection-scanner.js) inline their own pattern subsets for
hook independence and never called these functions.
Changes:
- Remove scanEntropyAnomalies + shannonEntropy from src/security.cts
- Remove scanEntropyAnomalies test block from tests/security.test.cjs
- Correct REQ-SCAN-INJ-02/-03 in FEATURES.md (EN/zh-CN/ja-JP) to
describe what actually runs live (injection patterns, invisible
Unicode) vs CI-only (base64-decode, codebase scan)
- Correct docs/security/baseline.md §2.4 to clarify live hooks inline
patterns, not import from security.cts
- Add regression test asserting the corrected contract
- scanForInjection retained: it serves as the CI codebase-scanner engine
VS Code is a net-new EoS runtime that — unlike every prior migration — is NOT
CLI-installed (Marketplace/VSIX extension). It has zero runtime==='vscode'
branches in bin/install.js and stays that way (regression-guarded); it is driven
entirely through the negotiated imperative Host-Integration adapter.
Registry + validator (the hard part):
- capabilities/vscode/capability.json (role:runtime): full hostIntegration block
(imperative / palette / active vscode.lm model / engine hook bus /
sandboxed-storage / mcp transport / sandboxed-web runtime; dispatch nested,
maxDepth 5 per VS Code's documented subagent depth).
- capability-validator.cjs extended so a role:runtime capability can legitimately
declare "extension-distributed, no config directory": new configHome.kind:'none'
+ installSurface:'none' (+ GATE-A pairing + the parity maps), with localConfigDir
and configHome.name made conditional on kind!=='none'. All 18 runtimes still
validate; getDirName returns a distinct sentinel (not '.claude') for a no-config
runtime.
- The add-a-registry-runtime tax: NON_INSTALLABLE_RUNTIMES exemption in the
runtime-flags drift guard, vscode added to global-config-home SPECIAL_CASED,
EXPECTED_PROFILES.vscode='ide', and the config-adapter/derivation/pin-count
guards updated. No golden-install fixture, model-catalog, or CONFIGURATION rows
(vscode never enters allRuntimes).
Dispatch + extension surface:
- Fixed vscode/extension.js's createHub()-no-args bug (every dispatch was
UnknownCommand, masked by a vacuous reachability test) — now reuses the shared
dispatchGsdCommand subprocess-shim (Node/desktop); the reachability test is
tightened to assert real dispatch.
- Promoted the #1933 host binding to a shipped vscode/host-binding.js; activate()
now composes the model/hookBus/stateIO seams through it. Corrected the model
seam to VS Code's real API (vscode.lm.selectChatModels() -> model.sendRequest();
vscode.lm.sendRequest does not exist) so the binding actually composes on real
desktop VS Code instead of throwing.
- New vscode/browser.js Web Extension entry with ZERO Node APIs (the engine's
config/capability loading is Node-bound, so the web entry registers the surface
and directs full dispatch to the native MCP server — honestly documented).
- UPGRADE 1: GSD skills as native Language Model Tools (contributes.languageModelTools
+ vscode.lm.registerTool), invoke() dispatching through the hub.
- UPGRADE 2: native subagent dispatch wired onto #runSubagent /
chat.subagents.allowInvocationsFromSubagents (fail-soft on API availability,
maxDepth 5 enforced).
- vscode/package.json: browser entry, engines.vscode ^1.105, chatParticipants +
languageModelTools contributions; fixed a stale activationPoints->activationEvents
manifest key. Added "vscode" to the package files array.
Docs (## vscode matrix section) + changeset (Added).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
audit-milestone's Nyquist scan classified a phase from `nyquist_compliant`
alone, so a phase seeded by plan-phase but never run through validate-phase
read PARTIAL — identical to a phase that validated and genuinely failed. The
template's `status` field could discriminate the two, but no workflow ever
promoted it off `draft`, so it was dead.
Make `status` live and read it:
- validate-phase.md §6: set `status: validated` in both the create (State B)
and update (State A) VALIDATION.md paths.
- audit-milestone.md §5.5: parse `status`; add a distinct NOT-VALIDATED bucket
keyed on `status: draft`, gate COMPLIANT/PARTIAL on `status: validated`, and
report `not_validated_phases` in the audit YAML.
- VALIDATION.md template: document the draft → validated lifecycle.
Tests & generated artifacts:
- Regression test folded into policy-138 (owning workflow-contract file);
fail-first verified vs origin/next (0 matches pre-fix).
- Regenerate golden-install-parity fixtures cleanly: adds the previously-missed
qwen.json and removes a contaminated `settings.local.json` entry that had
leaked into claude-local.json (the harness excludes hook-config files).
- Correct a stale validate-phase.md workflow-size-baseline entry.
Closes#2117
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Net-new EoS/pi installable runtime — purely additive (no prior runtime==='pi'
branches). pi is a bun-runtime programmatic-CLI whose /gsd command is registered
by a native ExtensionAPI extension and dispatches through the embedded engine.
Stage 1 (install plumbing):
- capabilities/pi/capability.json: full hostIntegration descriptor (imperative /
slash-programmatic / active-model / native-extension / bun) + hostBehaviors
{nativePlugin, pluginOnlyInstall}.
- --pi flag + interactive-menu renumber (All 17->18); pi added to RUNTIME_FLAG_IDS,
RUNTIME_LABELS, RUNTIME_META, allRuntimes/runtimeMap, model-catalog defaults.
- Install mirrors OpenCode: pi installs the gsd.cjs extension + the shared engine
payload (gsd-core + scripts + config markers) + the shared hooks bundle (spawned
by the extension at lifecycle events, like OpenCode's plugin). pluginOnlyInstall
EXCLUDES declarative command/agent/skill markdown, which pi has no host-read
surface for (its /gsd is programmatic). _installNativePluginIfDeclared (extracted
from the opencode-family path) copies pi/gsd.cjs -> ~/.pi/agent/extensions/gsd.cjs
(global) / .pi/extensions/ (local). pi added to package.json files.
- Golden: new pi.json (320 files: extension + engine + 27-file hooks bundle, no
markdown); the 16 other fixtures + claude-local change only by the shared
model-catalog hash line.
Stage 2 (real dispatch + upgrades):
- Shared dispatchGsdCommand() (shell-command-projection): bounded, no-throw
subprocess-shim to gsd-tools.cjs (the only full-surface dispatch path; no
in-process full-hub factory exists). Fixes pi/gsd.cjs's createHub()-no-args bug
(every dispatch was UnknownCommand) AND the identical bug in mcp-server.cts's
gsd_invoke_command, which a vacuous unknown-family-only test had masked (now has
a real dispatch regression test).
- pi/gsd.cjs: /gsd handler now (args, ctx) - tokenizes (quote-aware, via the
shipped hooks/lib/git-cmd.js) + dispatches real family/subcommand (not hardcoded
query/help); gsd_invoke gets a TypeBox (JSON-schema-fallback) parameters schema +
consumes params; getArgumentCompletions; before_provider_request active-model
steering (fail-open on null resolution); functional session_start /
before_agent_start / session_before_compact hook bridges (spawn the shipped GSD
hook scripts).
- EXTENSION_EVENT_SURFACES.pi expanded from ['tool_call'] to the full 30-event
vocabulary.
Docs (host-integration matrix + how-to) + changeset (Added).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The issue's "0 conditional branches" premise missed a live one: the `!isZcode`
shared-hooks exclusion (bin/install.js). Fold it onto descriptor-driven
hostBehaviors.skipSharedHooksInstall:true (zcode's golden has zero hook files —
byte-parity verified) and drop the now-unused isZcode destructure. Zero live
runtime==='zcode'/isZcode branches remain (AC2 source-grep guard over
bin/install.js + install-engine.cts + surface.cts + runtime-artifact-conversion.cts).
The 6 CLI-bookkeeping zcode mentions (--zcode flag, menu, roster, help) stay.
Reference test (declarative-reference-zcode.test.cjs): profileOf → declarative-cli;
createDeclarativeAdapter({runtime:'zcode'}).kind → declarative; a real install emits
the invocable nested-skills/commands/agents surface (no hooks); negotiateHostCapabilities
fail-closes (empty/corrupt descriptor; the nested/maxDepth undocumented sub-axes degrade
to most-restrictive); validateCapability clean.
UPGRADES documented as BLOCKED (verified doc gaps — NOT guessed, to avoid a
non-functional false-green): both of ZCode's documented capabilities lack a published
on-disk config format.
- Hook automation: zcode.z.ai/en/docs/plugin documents the Hook component only as
"automation hooks triggered on specific events" (capability detected from directory
layout) — no config file format/location/event schema. Cannot faithfully wire.
- MCP registration: zcode.z.ai/en/docs/mcp-services says servers are "stored in the
.zcode configuration file" (UI-only) with no documented on-disk filename/path/schema —
exactly the settings-filename gap the issue AC anticipated.
Both are documented (with the search trail) in the capability matrix + how-to, per AC4's
block-documentation clause; hookBus/transport stay declared for when ZCode publishes the
formats. No hook scripts or MCP artifacts added → no golden change, no other-runtime impact.
Golden: byte-identical for all 16 runtimes (the fold is byte-parity; no upgrade artifacts).
Matrix ## zcode EoS note + how-to; changeset (Changed). capability-registry regenerated.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fold all 10 residual isWindsurf branches in bin/install.js onto descriptor-driven
hostBehaviors (byte-parity — no fold changes any install output):
- 2 dead destructures dropped (uninstall, finishInstall); the dead
`else if (isWindsurf)` legacy agent-loop arm removed (windsurf ∈
_DESCRIPTOR_AGENTS_RUNTIMES → unreachable).
- skipSharedHooksInstall:true folds the two `!isWindsurf` shared-hooks exclusions.
- legacyDevinSkillsCleanup:true folds the `.devin`→`.windsurf` one-time cleanup gate.
- installsCommandBodiesForWorkflowDelegation:true folds the #1629 command-body copy
(workflow-delegation target — load-bearing; local-install verified intact).
- verificationStyle:"windsurf-workflows" folds the workflow-count report.
- Corrected stale _LEGACY_SCAN_SUBDIR_NAMES + hooks-json manifest comments (cursor + windsurf).
Zero live runtime==='windsurf'/isWindsurf branches remain across bin/install.js,
install-engine.cts, surface.cts, runtime-artifact-conversion.cts (AC2 guard scans all four).
UPGRADE (Cascade hook bus): wire GSD's write/command safety guards into Windsurf's
native hook bus. New hooksSurface 'windsurf-hooks-json' (VALID_HOOKS_SURFACES 7→8, GATE A
profile-marker-only allowlist, the HooksSurface union) + writeWindsurfHooksJson
(Cursor-templated, Cascade's flat {hooks:{<event>:[{command}]}} shape) writing
.windsurf/hooks.json with two BLOCKING pre-hooks:
- pre_write_code → gsd-windsurf-pre-write.js: blocks writes to a file outside the
active git worktree / into .git internals.
- pre_run_command → gsd-windsurf-pre-command.js: conservative destructive-command
deny-list (rm -rf of root/home incl. sudo/env/path-prefixed forms; fork bombs;
force-push refspec forms — HEAD:main, +main, --force/-f — to main/master/next).
Both use Cascade's protocol (stdin JSON, exit 2 + stderr to block, exit 0 to allow,
fail-open on error/timeout). Tokenize-based classifier (no catastrophic-backtracking regex;
4096-char cap) with the fail-closed false-positives fixed post-review.
The 4 advisory GSD guards + pre_mcp_tool_use + 5 post_* logging events are deliberately
NOT wired: Cascade has no context-injection channel for advisory hooks and GSD has no MCP
guard — porting them would be non-functional padding (documented; codebuddy #2098 / copilot
#2099 faithful-subset precedent). extendedHookEvents stays [].
Golden: the 2 guard scripts ship in the shared hook bundle (HOOKS_TO_COPY + the shared
managed-hooks-registry), exactly like cursor's 6 gsd-cursor-*.js scripts — so the 8
shared-bundle runtimes' fixtures gain the 2 inert windsurf scripts + the registry hash
(functionally inert for non-windsurf; the established cursor pattern). No install-output
change beyond that (the folds are byte-parity; skip-bundle runtimes untouched). New scripts
registered in managed-hooks-registry + build-hooks + INVENTORY. Tests: declarative-reference-
windsurf (adapter/axes/fail-closed + AC2 guard) + windsurf-hooks-bridge (live exit-2 blocking
+ allow/fail-open + ReDoS-bound + writer/reconcile/remove idempotency); VALID_HOOKS_SURFACES
pin updated to 8. Matrix hookBus delta + changeset (Changed). capability-registry regenerated.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Addresses findings from the orthogonal /code-review + /security-review passes:
- Markdown-injection (CRITICAL/HIGH): escape untrusted free-text (name, description,
license, author, eos axes) with mdInline() and size install/uninstall code fences
dynamically so a crafted entry cannot inject phishing links, break the summary
table, or escape the code fence in the committed, GitHub-rendered catalog.
- validateEntries no longer throws on a null/non-object array element (kept the
--json contract); rejects control characters in free-text fields; caps field
lengths and entry count; tightens the discussion and license regexes so neither
admits Markdown metacharacters / newlines.
- gen-registry treats a missing capabilities.json as an error (only eos.json is
optional pre-PR2); disambiguated from gen-capability-registry.cjs.
- Renders the required 'author' field (was captured but never shown).
- Adds tests for every fix: escaping/link-hijack/fence, null guard, control chars,
length + entry caps, tightened regexes, eos render path, interactions guards.
Refs #2182
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Implements the three pure functions in registry-schema.cjs (isValidGsdRange,
validateEntries, renderMarkdown) that were stubbed in the previous commit, turning
the red suite green: strict per-type schema validation (capability + eos), a
self-contained engines.gsd range validator (no semver dep), and deterministic
Markdown generation with shields.io release badges + per-entry Discussion links.
Regenerates docs/registries/capability-registry.md and adds the Added changeset.
Also hardens the isValidGsdRange fast-check property (letters-only major) so it
cannot intermittently generate a valid prerelease range and flake.
Closes#2182
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adds the discoverability-registry surface for issue #2182: JSON-sourced
capability/eos catalogs, a pure schema/vocab module (registry-schema.cjs)
with the ADR-857 loop points + ADR-1239 axes, thin validate/gen CLIs,
the registry-entry PR template, README spec, and CONTEXT.md glossary terms.
The three pure functions (isValidGsdRange/validateEntries/renderMarkdown)
are stubbed here so the comprehensive test suite fails first (red), per the
feature-implementation red-first directive; the next commit implements them.
Refs #2182
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fold Copilot's residual runtime-literal branches onto descriptor-driven
hostBehaviors. Several issue premises were inaccurate (verified via research)
and deliberately NOT followed: writesSharedSettings/"legacy exclusion list"
(per-runtime descriptor data, copilot's false is correct); RUNTIME_CONTENT_DISPATCH.copilot
+ installSurface==='copilot-instructions' (already descriptor-driven);
extendedHookEvents (closed Claude/Gemini enum — hooks extended in code instead);
reapply ternary (already folded, kimi #2095).
Real folds (all byte-parity — golden byte-identical for every runtime):
- src/install-engine.cts + src/surface.cts: the two `.agent.md` filename cutovers
(_copyStaged + _syncGsdDir) unified onto hostBehaviors.agentFileExtension via a
new exported agentFileExtensionFor() accessor (kills the two-mechanism divergence).
- src/runtime-artifact-conversion.cts: applyAgentPathRewrites' copilot skip →
hostBehaviors.noPathRewrite:true (antigravity #2096 precedent).
- bin/install.js uninstall: the two isCopilot cleanup branches → installSurface===
'copilot-instructions' gate (symmetric with install-time).
- bin/install.js: `!isCopilot` in the two skipSharedHooksInstall checks →
hostBehaviors.skipSharedHooksInstall:true (copilot has no shared gsd-*.js hooks).
- bin/install.js: three dead legacy inline-agent-loop isCopilot refs removed
(copilot ∈ _DESCRIPTOR_AGENTS_RUNTIMES → unreachable; byte-parity proven by clean
golden + real reachable-runtime install diffs). isCopilot dropped from 4 destructures.
Zero live `runtime==='copilot'`/`isCopilot` branches remain in bin/install.js,
install-engine.cts, surface.cts, or runtime-artifact-conversion.cts (AC2 guard scans all four).
UPGRADE 1 (multi-event hook bus): buildCopilotHookConfig() now emits preToolUse/
postToolUse/userPromptSubmitted/sessionEnd advisory handlers alongside sessionStart
(static inline bash/powershell — deterministic, golden-trackable). Only copilot.json's
gsd-session.json hash changes.
UPGRADE 2 (background dispatch): surfaced via the negotiated contract only —
dispatch.background:true exceeds the declarative-cli baseline and survives negotiation
with no downgrade warning. NO .agent.md frontmatter field (copilot has none). MCP
companion out of scope (AC4 names only 2 upgrades).
Tests: declarative-reference-copilot (adapter/axes/fail-closed + AC2 4-file source-grep
guard) + copilot-upgrades (live 5-event hook wiring; dispatch.background negotiation).
Matrix EoS note + how-to; changeset (Changed). capability-registry regenerated.
Incidental flaky-test RE-ARCHITECTURE (no-defer, maintainer-directed):
tests/opencode-review-reconstruction.property.test.cjs spawned ~600 synchronous
execFileSync('jq') subprocesses (numRuns:200 × 3 fast-check properties, one jq per
generated stream); a single jq freezing on a contended macos-22 CI runner hung the whole
unit-test chunk to its 600s kill (this PR's CI). --test-force-exit can't interrupt a
synchronous execFileSync, so the cure is to stop spawning per case, not just time-bound
it. Re-architected to run the SHIPPED jq program over the whole fast-check corpus in ONE
jq process: each generated stream is one compact-JSON array per line in a temp file,
`jq -c <PROGRAM>` (no -s) applies PROGRAM to each array (`.` == the array, exactly what
production's `jq -rs <file>` sees after slurping) and emits one result per line —
empirically byte-identical to the per-stream form across embedded-newline/empty/quote/
unicode/null-drop cases, and file-input (like production) so there's no stdin pipe to
deadlock on large I/O. ~600 spawns → 6; coverage unchanged (200-case corpus per property,
deterministic seeds) plus explicit boundary/diagnostic example batches. Still property-
tests the real shipped jq (no JS reimplementation). Per-call jq timeout retained as a
belt-and-suspenders bound.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fold CodeBuddy's residual runtime-literal branches onto descriptor-driven
hostBehaviors (the issue's "zero branches remain" premise was inaccurate — 2
lived):
- Folded the `if (isCodebuddy)` commands/ report duplicate (byte-identical to
the generic reportCommandsDir block above it) onto
hostBehaviors.reportCommandsDir:true, matching cursor's #2089 precedent.
- Deleted the dead `else if (isCodebuddy)` agent-conversion arm (codebuddy ∈
_DESCRIPTOR_AGENTS_RUNTIMES → gated out before the legacy chain, same dead-arm
pattern removed for augment/trae in #2097).
- Removed isCodebuddy from all 4 destructure sites (now comments only, mirroring
the #2096 isAntigravity fold). Zero live isCodebuddy reads remain.
UPGRADE 1 (extended hook bus): populate extendedHookEvents with the full
extended set SubagentStop/Stop/PreCompact/SubagentStart (qwen #2092 / kimi
precedent — codebuddy previously had extendedHookEvents:[] so it got NONE of
these). The generic applySettingsJsonHooks loop wires them, so CodeBuddy's
settings.json now gains all four subagent-lifecycle + stop/compact hooks it
previously lacked, matching Qwen/Kimi coverage. No source change (the
HOOK_EVENT_SURFACES SDK catalog is a locked dict; all 4 events are already in
the validator enum; hookEvents already 'claude'). settings.json is golden-
excluded → no golden change.
UPGRADE 2 (background dispatch): surfaced via the negotiated capability contract
only. CodeBuddy's dispatch.background:true legitimately exceeds the declarative-
cli baseline (false) and survives negotiation with no warning. NO agent-file
frontmatter field is emitted: the CodeBuddy CLI (GSD's install target,
~/.codebuddy/agents/) has NO background-dispatch frontmatter field — background
is a caller-side run_in_background invocation param (verified against
codebuddy.ai/docs/cli/sub-agents); the issue's agentMode/enabledAutoRun are
IDE-only (codebuddy.cn, a different product). Emitting them would be a non-
functional false-green, so it is deliberately not done.
Golden: byte-identical for all 16 runtimes (folds are console/failures-report
only; UPGRADE 1 → golden-excluded settings.json; UPGRADE 2 → no artifact
change). Tests: declarative-reference-codebuddy (adapter/axes/fail-closed
negotiation + AC2 source-grep guard) + codebuddy-upgrades (live install asserts
all 4 extended hooks wired; dispatch.background survives negotiation above
baseline). Matrix + install-on-your-runtime updated; changeset (Changed).
capability-registry regenerated.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fold augment's runtime-literal conversion branches onto descriptor-driven
hostBehaviors and delete dead code:
- Site A (_applyRuntimeRewrites case 'augment'): the 4 ~/.augment dot-dir
regexes now derive from getDirName('augment') via escapeRegExp (byte-
identical; getDirName('augment')==='.augment') — no runtime literal.
- Site B (applyRuntimeContentRewritesForCommandsInPlace): the
`if (runtime==='augment')` markdown-converter branch now reads
runtime.hostBehaviors.commandBodyConverter and dispatches through a local
COMMAND_BODY_CONVERTERS map (degrade-closed on unknown/absent name).
- Deleted dead `claudeToAugmentTools` map (zero refs; orphaned by ADR-1508
single-sourcing) and the unreachable `else if (isAugment)` agent-conversion
branch (augment ∈ _DESCRIPTOR_AGENTS_RUNTIMES → gated out upstream).
- Incidental orphan cleanup (no-defer): removed the equally-unreachable
`else if (isTrae)` agent-conversion arm left behind by trae's already-merged
migration #2094 (trae ∈ _DESCRIPTOR_AGENTS_RUNTIMES, same upstream gate).
copilot/windsurf/codebuddy arms are removed by their own pending migrations.
UPGRADE 3 (transport:mcp): register the GSD companion MCP server in Augment's
settings.json under mcpServers.gsd (Augment hosts MCP in settings.json, not a
standalone file). mergeGsdMcpServerIntoSettings mutates the in-memory settings
object finishInstall already writes (gated on hostBehaviors.mcpCompanion===
'settings-json'); non-destructive + idempotent; symmetric uninstall removal.
settings.json is golden-excluded, so no golden change. UPGRADE 1 (named/
background dispatch) + UPGRADE 2 (settings-json hook bus, Claude dialect)
were already live in production — this adds tests exercising both.
Golden: byte-identical for all 16 runtimes (folds preserve regex behavior;
MCP lives in golden-excluded settings.json) — verified by a real double-install
tree diff. Tests: declarative-reference-augment (adapter/axes/fail-closed/
undocumented-sub-axes + source-grep guard scoped to conversion-logic branches)
+ augment-upgrades (dispatch negotiation, hook-bus live install, MCP add/
idempotent/preserve/uninstall). Matrix + connect-gsd-mcp-server + a stale
install-on-your-runtime hook-ownership claim corrected; changeset (Changed).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
deriveProgressFromRoadmap read the `## Progress` table with two
4-column-only regexes, so every project past its v1.0 milestone — which
uses the 5-column milestone-grouped shape the same template ships
(| Phase | Milestone | Plans Complete | Status | Completed |) — parsed to
{completedPhases: null, totalPhases: null, totalPlans: null}. Because the
body is wrapped in a bare try/catch the failure was silent, and
phase.complete flipped the ROADMAP row to Complete but never advanced
STATE's progress percent / completed-phase count.
Parse the table by header instead: locate the `## Progress` section
(line-anchored h2, mirroring the writer's #2012 scoping so an earlier
Phase/Status/Completed-shaped table can't hijack it), read the header row
to find the Status, Completed, and plans columns by name, then index each
data row positionally. One code path now handles both the 4-column
greenfield and 5-column milestone-grouped shapes — the same understanding
the writer (cmdPhaseComplete, with its `cells.length === 5` branch)
already had. Preserves the 999.x backlog exclusion, M/N plan summing, and
the prior contract that a count of 0 is reported as null.
Regression tests fold into tests/state.test.cjs: the template's own
5-column Progress block parses non-null, both shapes derive identically,
999.x stays excluded in the 5-column shape, and decoy tables (an earlier
## Retrospective table and an ### Progress h3 subheading) do not hijack
the scope.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Fold all runtime==='kimi'/isKimi logic branches into descriptor-driven
hostBehaviors (localInstallDeferred, verificationStyle, agentManifestStyle,
reapplyCommand, doneBannerStyle) + add 'kimi' to _DESCRIPTOR_AGENTS_RUNTIMES.
Kimi's skills/kimi-agents dispatch was already descriptor-driven (converter-by-
name + kimi-agents kind). Zero isKimi/runtime==='kimi' branches remain.
UPGRADE 1 (native hook bus): new hooksSurface 'kimi-hooks-toml' + a marker-
delimited config.toml [[hooks]] emitter (buildKimiHooksTomlBlock/writeKimiHooksToml
in runtime-hooks-surface.cts; resolveKimiHooksTomlDir in runtime-homes.cts).
GSD's lifecycle hooks now wire into Kimi's native ~/.kimi/config.toml (Context7-
confirmed path) at SessionStart/PreToolUse/Stop/PreCompact/SubagentStart/
SubagentStop — kimi becomes a hooks/ consumer (the 3 && !isKimi exclusion guards
removed). config.toml holds absolute install paths so it's golden-excluded via
an exact relative-path (.kimi/config.toml), not a basename (which would blind
Codex's config.toml). New hooksSurface value added to the closed enum in
capability-validator + runtime-config-adapter-registry.
UPGRADE 2 (background dispatch): flip dispatch.backgroundDispatch true (Kimi's
Agent tool takes run_in_background; root agent already gets the Agent tool), so
negotiation no longer flattens dispatch. subagentToolkit stays 'undocumented'
per AC (coder/explore/plan have distinct tool policies).
MCP transport explicitly deferred (no installer-driven MCP for any runtime).
Golden: only kimi.json changes (hooks/ scripts now installed); all 15 others +
claude-local byte-identical (kilo/zcode keep their own exclusions). Tests:
kimi-imperative-reference (adapter/axes/fail-closed/hostBehaviors + source-grep
guard) + kimi-upgrades (config.toml [[hooks]] SessionStart + marker idempotency
+ backgroundDispatch negotiation). CONTEXT.md glossary + matrix + how-to updated;
changeset (Added).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fold trae logic branches into descriptor-driven reads: skipSharedHooksInstall
gates (dropped && !isTrae), and the case 'trae' path-rewrite arm now computes
the self-alias from the descriptor-driven dirName (.trae). Dead isTrae bindings
removed from uninstall/writeManifest/finishInstall. trae's skills dispatch was
already descriptor-driven (converter-by-name). RUNTIME_CONTENT_DISPATCH.trae is
left as a runtime-keyed table registration (its regex/callback rewrites can't be
a byte-identical descriptor map — matches cursor/windsurf/cline). trae stays in
RUNTIME_FLAG_IDS: isTrae still gates the agents-converter selection (agents out
of scope; removal gated on the cross-runtime agents-dispatch migration).
Byte-identical golden parity for all 16 runtimes.
UPGRADE: SOLO stage/trigger metadata — emitted Trae SKILL.md now carries
stage: workflow (descriptor-gated via hostBehaviors.soloStageMetadata) so
Trae's SOLO Agent can auto-invoke GSD skills at the corresponding stage. Field
shape is best-effort/inferred (Trae publishes no formal schema). trae.json
golden regenerated.
Tests: trae-imperative-reference (adapter/axes/fail-closed shouldFlattenDispatch
+ no runtime==='trae' source-grep, isTrae exempted for agents) + trae-upgrades
(stage: workflow on installed SKILL.md, descriptor-gated). Matrix note +
changeset added.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fold remaining isKilo logic branches into descriptor-driven reads:
finishPermissionWriter (uninstall cleanup), skipSharedHooksInstall (hooks
copy), and a skills converter-name registry (the artifactLayout.converter
field is now load-bearing, not decorative). frontmatterDialect stays the
documented dispatch key for frontmatter (no descriptor field for it). Dead
isKilo destructure bindings removed. Byte-identical golden parity for all 16
runtimes (opencode, which shares kilo's combined-family path, verified clean).
UPGRADE 1 (hook bus): install .kilo/plugins/gsd-core.js native plugin +
extensionEvents:"kilo" + EXTENSION_EVENT_SURFACES.kilo (OpenCode-fork bus).
UPGRADE 2 (active model): populate runtimeTierDefaults.kilo + thread
modelOverride through convertClaudeToKiloFrontmatter — model no longer stripped
from agents. UPGRADE 3 (MCP): document the gsd-core MCP companion under kilo's
mcp config key. UPGRADE 4 (named dispatch): agents/*.md mode:subagent roster is
the Task-tool dispatch surface (tested); subagentToolkit stays 'undocumented'
per AC so dispatch degrades to 'degraded' by design.
Model-catalog single-source edit ripples the shared model-catalog.json hash
into all 16 golden fixtures (expected). Inline defect fixes (no-defer): stale-
bake-guard resolveAgentDir 'agent'->'agents' (was a silent no-op for opencode/
codex), hardcoded 'Removed OpenCode plugin' uninstall log -> generic, and the
connect-gsd-mcp-server.md OpenCode mcpServers->mcp doc error.
Tests: kilo-imperative-reference (adapter/axes/fail-closed/degradation/
hostBehaviors + widened isKilo source-grep across 4 modules) + kilo-upgrades
(plugin parity+load, model-override converter, agents dispatch surface, MCP
doc). Matrix + how-to + config docs updated; changeset added.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Record resolved UI-state considerations in the UI-SPEC, backward-compatibly
(Phase 2, WIRE-02). templates/UI-SPEC.md gains a '## UI Considerations' section
(analog of SPEC '## Edge Coverage') after '## Copywriting Contract' — a
| Category | Element(s) | Status | Resolution / Reason | table with
covered/backstop/unresolved rows in the locked probe-core projectTruths format
the shipped plan-phase lift (plan-phase.md:921) reads. Empty/error COPY stays in
Copywriting; this section covers shape-rooted STATE and references those rows
(de-dup).
Tests: docs-fixtures parsed-heading assertion (UI Considerations present +
distinct from Copywriting Contract; allow-test-rule, parsed structure); typed
backward-compat (projectTruths(undefined/[])===[], old UI-SPEC still plans —
Hyrum), format-match, and idempotency (proposeElements determinism). The
template-structure test was the RED driver.
Install-parity cascade (INVENTORY-MANIFEST + 16 golden fixtures +
agent-size-baseline) deferred to Phase 3 SHIP-01, as planned.
Claude-Session: https://claude.ai/code/session_01BKt4hgNZwXSeJYJtYAQUSS
Add the live ui-phase producer path for the UI-consideration probe (Phase 2,
WIRE-01). Two small exports on the Phase-1 adapter — proposeElements (the
propose-then-confirm view of detected kinds + applicable categories) and
autoResolve (the deterministic --auto floor that never dismisses and never
auto-backstops an unclassified item, #1110) — plus a post-verification
'## 9.5 UI-Consideration Probe' step in ui-phase.md mirroring spec-phase 5.5's
RUNTIME_DIR shim + fatal-invoke/malformed-report/zero-applicable fail-closed
guards, propose-then-confirm (the partial-cue recall mitigation), and the
'## UI Considerations' write-back in the shipped plan-phase.md:921 lift format.
autoResolve is the CODE floor; the covered-upgrade stays workflow prose (the
two-layer --auto). Un-upgraded backstops route to insufficient_spec ->
human_needed at verify, never a silent pass (#1154).
Tests: +8 typed (proposeElements shape/determinism, autoResolve never-dismiss,
partial-cue strict-subset) — structured-value only. ui-phase.md size baseline
ratcheted 15477->24447 (under DEFAULT cap). The plan-phase.md PRE_PHASE6 ceiling
stays RED pending #1852 (unchanged from Phase 1).
Claude-Session: https://claude.ai/code/session_01BKt4hgNZwXSeJYJtYAQUSS
Reference doc mirrors edge-probe.md structure but links rather than re-argues;
states the MIXED-axis boundary (closed compiled shape-rooted subset here; open
UX subset - real-time/offline, a11y depth, i18n/RTL - prose-owned in
domain-probes.md). Docs-parity test pins doc taxonomy ids == code UI_TAXONOMY
ids and asserts disjointness from domain-probes.md topics (ADR-456
runtime-contract exemption, see #1867).
Third probe-core adapter on the UI element/state axis, mirroring edge-probe.
Closed 8-id shape-rooted UI_TAXONOMY + element-cue relevance filter
(UI_CUES -> classifyElement -> applicableCategories); unclassified fail-loud
soft-signal (#1110); fail-closed on invalid authored element kinds. All
lifecycle/merge/validation delegated to probe-core verbatim (no fork). Item
question carried in the shared Item.probe field. LIFT-01 proven at the
probe-core primitive level (projectTruths/dispositionForUnverifiableTruth):
backstop considerations route to insufficient_spec, never a silent pass.
Tests assert typed returns off the built .cjs (no source-grep).
#2056 fixed the foreign-prefix collapse for init plan-phase only.
The identical defect remained in three sibling commands that called
findPhaseInternal/getRoadmapPhaseInternal without the guard:
- cmdInitExecutePhase
- cmdInitVerifyWork
- cmdInitPhaseOp
Extracted the #2056 guard into shared helpers (guardedFindPhase /
guardedGetRoadmapPhase) and routed all four init commands through them.
Deleted the local parsePhasePrefix/isForeignPrefixedPhaseQuery copies in
init.cts — the canonical export from phase-id.cts is now used directly,
eliminating the drift risk flagged by both reviewers.
Added 5 regression tests (3 reject + 2 accept-branch) mirroring the
#2056 plan-phase tests.
Update runtimeTierDefaults.codex and providerPresets.openai in
model-catalog.json to the GPT-5.6 family (gpt-5.6-sol/terra/luna),
advancing from the superseded GPT-5.4/5.5 generation.
Model IDs verified against OpenAI developer API docs:
- gpt-5.6-sol: flagship, /, reasoning xhigh
- gpt-5.6-terra: balanced, .50/, reasoning medium
- gpt-5.6-luna: fast/cheap, /, reasoning medium
Tier mapping is 1:1 (Sol↔flagship, Terra↔balanced, Luna↔fast),
so profile semantics are unchanged — only the underlying IDs advance.
Updates: catalog JSON, test assertions (catalog defaults), docs
(CONFIGURATION.md + zh-CN/pt-BR translations, workflow settings),
and changeset.
Closes#2122
Final convergence review found the shared <tag> seam introduced 3 behavior
regressions; fixed all + locked with tests:
- #557 REGRESSION: stripTaggedBlocks's attribute-tolerance stripped `<details open>`
(the ACTIVE-milestone marker) that the old `<details>`-only regex preserved. The
seam now takes `allowAttributes` (default false) — details/decisions strip is
attr-INTOLERANT (preserves `<details open>`); only `<task type="…">` opts in.
Regression test added to roadmap-parser + markdown-sectionizer suites.
- verify.cts actionZones (negative-grep-echo security scan): reverted to a bounded
to-first-close scan `<action>([\s\S]{0,20000}?)</action>` so a grep-echo trick
can't hide behind an unterminated inner <action> (the seam's stop-at-next-open
would drop it). ReDoS-safe via the cap.
- check-command-router HTML-comment strip: `(?:-->|$)` fallback wiped to EOF
(fail-closed spurious gate block) — replaced with stop-at-next-open so an
unclosed `<!--` leaves downstream tags intact.
- Updated the extractTaggedBlocks nested-tag tests to the new (stop-at-next-open)
behavior: `<x><x>inner</x></x>` -> ['inner'].
All vectors still linear; every fix verified in-process.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Review caught that the prior commit bounded only the paren tag clause and left
the SIBLING bracket-prefix `(?:\[[^\]]+\]\s*)?` (same host regexes, before Phase)
UNBOUNDED — the identical quadratic reachable via a `[...]` run (measured ~16s at
1.7MB). Bound `[^\]]+`/`[^\]]*` -> {1,200}/{0,200} across all 19 phase/milestone
heading prefixes. Comprehensive re-measurement now shows EVERY vector linear
(bracket/paren/id/name/milestone all ~2-44ms at 2.45MB; bracket scaling
2k->2ms, 4k->5ms, 8k->10ms). Also: update the #1729 literal-mirror parity test
off its stale unbounded constant, and add limit-1 (199) boundary coverage.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The canonical OPTIONAL_PHASE_TAG_SOURCE tag clause `(?:\s*\([^)\n]*\))?` (and its
inlined literal mirrors across 11 modules) had an UNBOUNDED body, making the
optional-group + /g header scan quadratic on adversarial ROADMAP.md/STATE.md — a
long run of `(` after a header ran ~18.8s at 1.7MB. Bound the body to {0,200} in
the constant AND every mirror in lockstep (the #1729 "both forms change together"
contract), so the scan is linear: the same 1.7MB input now resolves in ~9ms
(measured), while real tags (a handful of chars) still match and a 201-char tag
is rejected. Added a #2128 boundary regression to the #1729 suite.
Pre-existing (byte-identical before/after the Phase 4 migrations); folded in at
maintainer direction rather than deferred.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Re-review found the `// phase-id-owner:` suppression treated a `//` embedded in a
string literal as a comment — help/doc text quoting the sanction syntax (the exact
string the scanner's own main() prints) would silently suppress a real
re-derivation. Require the marker to LEAD its own comment line (`^\s*//…`), so a
`//` inside a string or trailing a code line never counts. All 5 real sanctions
are already dedicated lines (scanRepo stays green); trailing same-line sanctions
are no longer honored — put the comment on the line directly above.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Correctness review of the Phase 4 guard found the allowlist over-broad and the
scanner/guards evadable. Fixed all findings:
- Migrate 9 sites that were wrongly sanctioned: their regex is the PURE canonical
token (`\d+[A-Z]?(?:\.\d+)*`, no variant), byte-identical to already-migrated
siblings. The old justification argued against swapping to the extractPhaseToken()
FUNCTION (behavior-risky) — but the guard only wants the same regex built from
the SOURCE string (byte-equal, zero risk). Coverage is now 32 migrated / 5
sanctioned, not the overstated 23 / 14 (audit.cts x3, uat.cts, init.cts x4,
roadmap-upgrade.cts). Each conversion proven byte-equal (.source + .flags).
- Harden the drift detector: also catch the `[0-9]`-in-place-of-`\d` variant;
document the accepted limits (cross-line split, semantic restructuring —
covered by the identity guard + review, not a text scan).
- Sanction robustness: a `phase-id-owner:` marker now counts only inside a `//`
comment (a bare substring in a string no longer suppresses a real flag), and
the preceding-line window skips blank lines (an auto-formatter's blank line no
longer reactivates the flag).
- roadmap-parser.cts:462 comment: corrected — that regex carries no /i flag, so
its [A-Za-z] class does real case work (matches state.cts:1409's rationale).
- Identity guard: surface require failures instead of silently skipping, and
floor coverage at >75% of consumer modules (inspects 156/157).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 4 of epic #2121 (ADR-2121 Decision 7), closing the recurrence loop that
produced #2111 / #2114 / #2104: no module outside src/phase-id.cts may
re-implement phase-ID parsing without failing CI.
- phase-id.cts: add PHASE_NUMBER_TOKEN_SOURCE — the canonical phase-number-token
grammar (\d+[A-Z]?(?:\.\d+)*) for enumeration/scan call sites, the ANY-phase
counterpart to phaseMarkdownRegexSource(n)'s known-number lookup. Extend-only
(never touches normalizePhaseName; blast radius 79 fns / CRITICAL).
- scripts/lint-phase-id-drift.cjs: pure findPhaseIdRegexDrift(text) + scanRepo(root),
wired to `npm run check:phase-id-drift`. Flags a literal re-derivation of the
canonical token (both /\d/ and new-RegExp `\\d` escaping, plus the [A-Za-z] and
[.-] near-variants) anywhere in src/** outside phase-id.cts, unless sanctioned
with `// phase-id-owner: <reason>`. Narrow by design: bare \d+, digits-only
captures, \w ids, status-message text and pipe-tables are not flagged.
- tests/phase-id-drift-guard.test.cjs: fail-first drift cases (AC1) + live
scanRepo(ROOT) zero-drift (AC3) + identity guard — phase-id.cjs exports the
complete locked surface and no consumer re-exports a divergent copy (AC2).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Re-review found the test comment + changeset prose inaccurately claimed a bare
query "always" surfaced malformed_roadmap. Empirically, on origin/next a
project-code-prefixed checklist entry was a silent {found:false} for BOTH query
forms — the prefixed pass discarded its malformed candidate and the bare regex
could not match the PROJ- prefix at all. The unified 3-source lookup newly grants
the diagnostic to both forms; correct the prose to say so. No logic change.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Adversarial review of the Phase 3 branch surfaced three verified defects; fix
all three in place (no defer):
- install-runtime-artifacts.test.cjs: finish the fold-triplication dedup started
earlier (only enh-1511 had been collapsed). 11 B1-batch __foldDescribe blocks
were byte-identical triplicates (~5.9k lines, ~49% of the file), tripling the
subprocess-spawning installer suites under --test-concurrency — the same
starvation that produced the temp-dir races this branch fixes. Byte-identity
verified per block before removal; 230 distinct test/it titles preserved
(origin/next: 230 -> 230), interleaved B3/B5/B6 singletons untouched.
- config-get-default.test.cjs: make runExpectError faithful to production. The
throwing process.exit seam was caught by cmdConfigGet's "No config.json"
guard and reclassified into a spurious 2nd error() with the wrong reason
(CONFIG_PARSE_FAILED). Drive io.setJsonErrorMode + carry the original message
on the sentinel so the guard re-throws (single fire), assert exitCount===1,
and strengthen both probes to assert the typed reason (CONFIG_NO_FILE /
CONFIG_KEY_NOT_FOUND).
- roadmap.test.cjs: lock the #2121/#2114 malformed_roadmap parity — a
project-code-prefixed query against a checklist-only roadmap now surfaces the
same diagnostic a bare query always did (fails on prior silent-empty behavior).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
The prohibition-enforcement real-runner tests linted src/clock.cts (a .cts) as
their clean target. Under eslint.config.mjs's type-aware block for src/**/*.cts
(recommendedTypeChecked + parserOptions.project: tsconfig.build.json), each eslint
spawn loaded the WHOLE tsconfig.build.json program (~2s, CPU-heavy). The
real-runner tests spawn eslint repeatedly; under --test-concurrency those
full-program type-checks oversubscribed the bench CPU and blew the 60s subprocess
bound -> fail-closed (intermittent, load-dependent — passed 24241/24241 in an
earlier run, failed here).
Root fix (not a retry/timeout bandaid; measured projectService = no faster since
a single-file .cts lint still loads type info): add tests/_ff_lint_clean.cjs, a
KNOWN-CLEAN lint-scoped .cjs companion to _ff_lint_violation.cjs, with a
flat-config block enabling local/no-source-grep so the clean pass stays
non-vacuous. Repoint the 6 src/clock.cts real-runner usages (5 targets + the FF-02
toothless violationFixture) at it. Each spawn is now ~0.8s non-type-aware (no
whole-program load) — starvation removed. All 6 tests' semantics verified
in-process (SF-01 greens; toothless/fail-closed stay unverified); full-repo
`eslint .` green.
Refs #2126, #1259
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 3's gsd-test surfaced 8 pre-existing test-isolation races (in #2090's test
files now on next). Per CLAUDE.md's no-defer rule these are fixed inline in the
current change. Root-caused via /qa-test-architect — all bad-test (the
rewrite-engine production code is race-free):
- install-runtime-artifacts.test.cjs: the "rmSync when readFileSync throws" test
diffed the SHARED os.tmpdir() for gsd-cmd-rewrites-* dirs and force-deleted any
new one with no ownership check. Under --test-concurrency it deleted a sibling
test file's LIVE tempDir mid-copy (the #1575 "ENOENT .../graphify.md") and
misattributed it as its own leak. Fixed: capture the exact tempDir THIS call
creates (fs.mkdtempSync monkeypatch, restored in finally) and assert only on
that — never sweep/delete the shared os.tmpdir(). Also deduped the enh-1511
block the #1969 consolidation folded in 3x byte-identically (#1970/#1974/#1975)
down to 1 copy; 308 unique test titles unchanged (verified).
- issue-1575-agent-descriptor-parity.test.cjs: a missing }); nested the M2
'cursor attribution' test inside the per-runtime loop so it ran 7x (widening
the tempDir window). Fixed the brace -> runs once as a describe sibling.
- config-get-default.test.cjs: local run()/runRaw() spawned node via
execFileSync with a fixed 5s timeout and no retry -> ETIMEDOUT under Docker
load. Redesigned to call cmdConfigGet in-process (fs.writeSync fd-capture +
process.exit sentinel, both restored in finally) — no subprocess, no wall clock.
- runtime-artifact-conversion.cts: fixed the stale "No production caller today"
JSDoc on rewriteStagedCommandBodies (real callers: applySurface,
createRuntimeArtifactInstallPlan) — the false doc invited the bad test.
Refs #2126, #2090
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Phase 3 of epic #2121. cmdRoadmapGetPhase and getRoadmapPhaseWithFallback now
iterate the shared roadmapPhaseLookupSources (exact -> numeric -> prefix-tolerant,
owned by phase-id.cts since Phase 1) instead of a hand-rolled 2-source lookup, so
all three roadmap resolvers share one resolution contract.
Drives #2114: `roadmap get-phase <bare-N>` now resolves a drifted
`### Phase AB-29:` heading (matching getRoadmapPhaseInternal / init.phase-op),
previously EMPTY from the CLI. The malformed_roadmap checklist-fallback and the
milestone-then-full precedence are preserved (a milestone checklist never blocks
a full-roadmap header match).
Behavior reversal (approved in-session): a bare query now also resolves a
*drifted-only* prefixed heading when no bare sibling exists, reversing the #3599
counter-test's expectation. #3599's real anti-steal intent (a bare sibling wins
over a distinct prefixed one) is preserved by the exact->numeric->prefix-tolerant
ordering and re-asserted in the updated test; a new #2114 block covers the
drifted-only case. Fail-first demonstrated.
Closes#2126
Refs #2121
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
