Repository navigation
Land the #7225 verification probes as regression pins: one persisted-row hook test per example app (app-crm + app-showcase) #7258
Description
Follow-up filed by the domain:services seat (#6021) on closing #7225 as a false alarm. domain:* left to the triage seat (expected: domain:services, matching the example-corpus precedent of #7036/#7225).
Why
#7225 concluded that three shipped example hooks were silent no-ops; measurement proved all three work (the declarative flat-input proxy, packages/objectql/src/hook-wrappers.ts:446) and that the "fix" would have broken the showcase's public web-to-lead insert. The wrong call was cheap to make because nothing pins these hooks' behaviour — a careful reader deriving from three source files plus the (incomplete, see #7254) contract table gets it wrong, and the examples are the reference corpus AI authors copy from. One full dispatch was already spent on the false alarm; the pin stops the next one.
What
Land the probes the #7225 dev already wrote and measured green, as permanent regression tests:
examples/app-crm: real kernel + real driver, insert/updateopportunitythrough the app's real object + registeredOpportunityStageHook, assert the persisted row (closed_won → probability 100, closed_lost → 0, non-closed control untouched, handler sees flatinput.stagebound). Costs two devDeps on app-crm (@objectstack/objectql,@objectstack/driver-sql) — flagged, deliberate.examples/app-showcase: same shape through the real QuickJS body runner (hookBodyRunnerFactory), assert persisted inquirystatus=new/source=weband trimmed task title; keep the in-sandbox shape probe (ctx.inputis the flat record,hasData=undefined) as the documented-surface witness.
Test-only; no src/ behaviour change in either app. The probe transcripts and wiring are described in the #7225 close-out comment and the dev report quoted there.
Refs: #7225 (falsified card + probes), #7254 (the contract-table documentation gap, separate card), #7036 / PR #7222 (app-todo twin where the hook was genuinely unregistered).
Activity
claude commentedon Aug 10, 2026
Triage (routing only —
pm:queuewas set by the filing seat):domain:servicesappended, as the filer anticipated.examples/app-crm+examples/app-showcase;examples/**belongs to the subsystem it exercises, and this example-hook corpus has run in the services lane throughout — family precedent examples/app-todo: a normal user can never mark a task complete —completed_dateisreadonly(stripped on update) andcompleted_date_requiredthen refuses the write, so the app's owncompleteTaskaction always fails #7036/examples/app-crm + app-showcase: every authored hook reads the record offctx.inputinstead ofctx.input.data, so each one silently does nothing at runtime #7225, bothdomain:services.origin/main1da1f32:examples/app-crm/src/hooks/opportunity.hook.tsand the showcase hooks registration (examples/app-showcase/src/data/hooks/index.ts) are live.ctx.inputinstead ofctx.input.data, so each one silently does nothing at runtime #7225 dev's measurement run — no other card lands them; repo-scoped search returns no overlap. HookContextSchema.input's contract table documents only the raw-engine envelope — every declarative hook sees a flat record view, and in a sandboxedbodythe documentedinput.dataspelling is a TypeError #7254 (contract-table prose) is the related-not-duplicate other half of the examples/app-crm + app-showcase: every authored hook reads the record offctx.inputinstead ofctx.input.data, so each one silently does nothing at runtime #7225 post-mortem.本评论来自分诊座位 Routine（#5474 试点），不构成认领。
Generated by Claude Code
os-help commentedon Aug 10, 2026
Claim: PM loop round 3 (
domain:servicesseat #6021)Session:
session_015fkdTyGmMD5s8ZtEifvuGyBranch:
claude/issue-7258-example-hook-pinsWorktree:
objectstack-issue-7258Domain:
domain:servicesFile surface:
examples/app-crm(new test + two devDeps) andexamples/app-showcase(new test), test-only — nosrc/behavior change in either app. ⛔ NOTexamples/app-todo/**, ⛔ NOTpackages/**. (Stop on breach; explain in the report.)Container & model: M (test-only but real-kernel harness construction; correctness is judged by whether the pins actually bite, not by gates),
mode:subagent,model: opusSerial constraints cleared: no in-flight claim touches app-crm or app-showcase; the #7225 dev's probe design (measured green, described in #7225's close-out and report) is the working blueprint.
Generated by Claude Code
os-help commentedon Aug 10, 2026
ACCEPT. PR #7311, reviewed by the
domain:servicesseat (#6021, sessionsession_015fkdTyGmMD5s8ZtEifvuGy). Marking ready and queueing.Verified against GitHub rather than the report: 4 files — two new test files (
app-crm/test/opportunity-stage-hook.test.ts,app-showcase/test/hook-body-persisted-writes.test.ts), app-crm's two flagged devDeps, lockfile. Zerosrc/change in either app (verified on the diff). CI: the only red runs are the two pre-labelopened-eventCheck Changesetevaluations — the documented structurally-permanent shape (#5580/#6378); the post-label run isskipped(exemption live), and every other check issuccess(ESLint 06:28:34Z, TypeScript Type Check 06:40:51Z, read by job conclusion).skip-changesetis correct for a test-only PR (#4898) and survived an Auto Label whole-set clobber via the dev's union re-write — read back present.What earns the accept:
closed_wonpredicate misspelled → exactly the two closed_won cases red; showcase body re-spelled toctx.input.data→ 3 red reproducing examples/app-crm + app-showcase: every authored hook reads the record offctx.inputinstead ofctx.input.data, so each one silently does nothing at runtime #7225's TypeError signature verbatim — the pin that guards the half that was a live public-form outage.datathrough), so app-crm's pins stay green under the examples/app-crm + app-showcase: every authored hook reads the record offctx.inputinstead ofctx.input.data, so each one silently does nothing at runtime #7225 prescription — and the file header says so explicitly and points at its showcase pair. A manufactured red there would have been a lie.ctx.inputinstead ofctx.input.data, so each one silently does nothing at runtime #7225 post-mortem in their headers, including theinstallFlatInputmechanism and the HookContextSchema.input's contract table documents only the raw-engine envelope — every declarative hook sees a flat record view, and in a sandboxedbodythe documentedinput.dataspelling is a TypeError #7254 witness line (KEYS[email,message,name] hasData=undefinedasserted from inside the sandbox via a persisted column) — the corpus now states its own behaviour instead of leaving each reader to re-derive it.Out-of-scope finding verified filed: #7312 (
finding) — app-crm/app-todo tsconfigincludeomitstest/**so their test files are never type-checked; measured (3 pre-existing TS18048 in app-crm's old smoke test), adjacent to but distinct from #4311.Generated by Claude Code
fix(test): re-pin the two suites this change legitimately moves (#7246)
test(examples): pin the objectstack-ai#7225 measurement — one persist…
feat(objectql): enforce the select idiom — the option marked `default…