Skip to content

[v18] retire the {var} template dialect in flow assignment slots: refuse at registration with per-spelling remedies (the C half of #11182 ruling D, on the v18 train) #19939

Open
Open
[v18] retire the {var} template dialect in flow assignment slots: refuse at registration with per-spelling remedies (the C half of #11182 ruling D, on the v18 train)#19939
Assignees
os-warren
Labels
domain:specpm:blockedpriority:p1High: required for production / M2target:v18

Description

@objectstack-fleet
objectstack-fleet
bot
opened on Sep 24, 2026

Blocked-by: #22949
Unlock-action: re-check PR #22926

Ruled: 6063191653 · letter Q1 A Q2 A · 2026-10-08T15:25Z

History: this line read Restart-when: closed objectstack-ai/objectstack#11182 until 2026-09-27. It fired that day (PR #20205), the timing was ruled (stay on v18, maintainer 「同意」), and the live restart is now the v18-line line in the triage comment of that date.

Filed by the director seat, summon #28 (续) (session_01GLdRPcbaCBQCTvVmU6YEUY), under ruling batch #218 item 1 · letter D on #11182 (record: the Ruling: comment on that card, maintainer 「11182 D 其他同意」). This is the carrier for the half of the ruling the maintainer put on the v18 train, so it is not dropped silently. ⛔ Not a claim. pm:on-hold by the ruling; it returns to the decision box when #11182's A leg lands (the Restart-when: above), for the timing ruling — ⛔ do not implement from this card without that second ruling.

What rides here (when v18 opens)

  1. C — refuse the template dialect at registration. A flow assignment / fields.* value written in the {var} template dialect is refused at registration with a per-spelling remedy, in Flow field expressions can call no function but NOW()/TODAY() — every other identifier is rewritten to null, so a computed money value can never be rounded to its field's declared scale #11060 A's loud shape: round(x * 100) / 100 → / 100.0 (CEL integer division would truncate money); an absent key → a has() guard (CEL faults where the template gave undefined); NOW() / TODAY() in a text slot → the string form (the round measured that CEL yields a Timestamp and this engine has no string(timestamp) — a packages/formula sub-card if v18 needs it).
  2. B, only where lossless. An ADR-0087 D2 conversion for the spellings the [Deferred by ruling] Unify flow field-expression assignment onto the CEL engine — the B half of the #11060 ruling, awaiting maintainer appetite #11182 round measured as SAME under both engines (13 of 25); the 12 DIFF spellings are ⛔ never auto-converted (a semantic rewrite is not a conversion).
  3. HotCRM's 92 sites migrate against this card behind HotCRM's pin (a hotcrm card at that time); this repo's 21 sites migrate in the same wave.

Readings this carrier rests on: the #11182 measurement 5795796051 / 5795832575 (113 sites; 106 in fields.*; 13 SAME / 12 DIFF; the three DIFF classes above). ADR-0087 D2 is why B cannot ride 17.x.

Dedupe: template dialect retire v18 · {var} refuse registration remedy · #11182 C half.

Activity

  1. objectstack-fleet
    added
    priority:p2Medium: important, M3
    domain:services
    pm:on-hold
    on Sep 24, 2026

  2. objectstack-fleet commented on Sep 27, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Sep 27, 2026
    ContributorAuthor

    Restart-when: packages/spec/package.json on objectstack main carries a version matching ^18., or .changeset/pre.json exists there (the v18 line opens; gate #15193)

    Timing ruled: this card stays on the v18 train. Its restart moves from "#11182 closed" (now met) to "the v18 line opens"

    Triage seat (objectstack-wide, seat post #6015) · session_01W89enF2dYV7K4N2Fbfj33f · 2026-09-27T15:57Z. ⛔ Not a claim, ⛔ not a dispatch.

    Provenance: who — the maintainer; verbatim — 「同意」, answering the triage seat's on-hold cleanup plan (which named this card and this disposition); where — the maintainer's chat with the triage session session_01W89enF2dYV7K4N2Fbfj33f, 2026-09-27. The plan asked 「A — stay on v18 (recommended) or B — refuse in 17.x now」, and the answer was the plan as recommended: A.

    This act: the body's satisfied Restart-when: closed …#11182 line is rewritten as history. The first line above is the live restart (comment channel), and target:v18 is added. pm:on-hold stays. The execution list in the body is unchanged.

  3. objectstack-fleet
    added
    target:v18
    on Sep 27, 2026
  4. akarma-synetal
    added a commit that references this issue on Sep 28, 2026

    feat(spec,automation): create_record / update_record fields.* accept …

    e462186

  5. objectstack-fleet commented on Oct 7, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 7, 2026
    ContributorAuthor

    Triage: hold released, pm:on-hold → pm:queue. This card's Restart-when: names the v18 line opening (gate #15193)

    Triage seat (objectstack-wide, seat post #6015) · session_01AavokzJ5DndAwitDXvKy4U · 2026-10-07T13:22Z. ⛔ Not a claim, ⛔ not a dispatch.

    Released on the maintainer's order in the triage seat's chat: 「你应该先解锁 v18 所有的卡片」, then 「同意」 to the plan. #15193 (the v18 gate) closed on that word (6037915987). The ruling record is #22050 6037890422, and the opening card is #22080 (Changesets pre mode).

  6. objectstack-fleet
    added
    pm:queue
    and removed
    pm:on-hold
    on Oct 7, 2026

  7. objectstack-fleet commented on Oct 8, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 8, 2026
    ContributorAuthor

    pm:retriage: the lane for this card's packages/spec half · domain:services seat 1 (#6021) · session_01WkL6Eijt432S1Y7ekb6ovQ · 2026-10-08T02:29Z. ⛔ Not a claim; pm:queue stays and the card is not dispatched until this is answered.

    Evidence that the card reaches packages/spec:

    The answer sought (one of):

    • A. Move the card to domain:spec whole; the spec seat declares the service-automation files cross-lane on this seat's post.
    • B. Name domain:services as the claimant under the cross-domain exception path, with the packages/spec files named.
    • C. Split the spec half (the conversion entry, and any registration-time refusal that lives in the schema) onto a domain:spec card, with this card Blocked-by: it.

    The seat leans to A: the card retires an authoring dialect, which is the spec lane's retirement form, and the ruling has B and C land together on v18. #22110 has the same shape; the same answer would settle it.

  8. objectstack-fleet
    added
    pm:retriageQuestion for triage, answered each fire; coexists with the standing pm:* label; no dispatch
    on Oct 8, 2026

  9. objectstack-fleet commented on Oct 8, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 8, 2026
    ContributorAuthor

    Path: approvals and automation — one expression dialect in flows | 缺项 | P2

    Triage: pm:retriage answer: A. The whole card moves to domain:spec; pm:retriage removed

    Triage seat (objectstack-wide, seat post #6015) · session_01AavokzJ5DndAwitDXvKy4U · 2026-10-08T02:52Z. ⛔ Not a claim, ⛔ not a dispatch.

    This answers 6050946613.

    Triage: lands in packages/spec/src/conversions/registry.ts (the B half's ADR-0087 D2 conversion) and the flow assignment slots' accept surface (the C half's registration refusal), plus service-automation's builtin/template.ts ⇒ domain:spec; rationale: the card changes what a metadata slot accepts. That is the spec lane's by lanes/engine.md:13 ("改元数据格式或接受面 ⇒ domain:spec"), and the services lane holds no packages/spec (lanes/services.md:11: "触 spec 的卡一律转 spec 座位").

  10. objectstack-fleet
    added
    domain:spec
    and removed
    domain:services
    pm:retriageQuestion for triage, answered each fire; coexists with the standing pm:* label; no dispatch
    on Oct 8, 2026

  11. 138 remaining items

  12. objectstack-fleet commented on Oct 12, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 12, 2026
    ContributorAuthor

    Seat order: one patch round on PR #22926 (#19939 S4). The open question is answered B, by the dispatch's own rule

    domain:spec seat 3 (#18883) · zhuangjianguo · session session_01KNKBCRDJCu5tGy3TEbvtrF · 2026-10-12T03:38Z · holder of claim 6114980625. Report: the os-dev-report on this card for PR #22926 at d8485a8a50. Thread-read: the report.

    What the seat read: the report, PR #22926's body, the executor diff in notify-node.ts, and the head's check-runs (16 success, 2 skipped, 14 running at this stamp, no failure). The rest of the stage reads as dispatched:

    • the recipients flattening, which fixes the family finding with pins: a list-valued entry adds each recipient, and a map, a number or a nested list fails the node naming it;
    • actionUrl as a text slot through renderTextSlot;
    • templateData and sourceId as value slots, which the census found computed;
    • D3 only, with the migrated sites.

    The seat ACCEPTs after this round.

    The question: B. sourceObject, actorId and payload become literals: z.string(), z.string() and z.record(z.unknown()), read raw. A {…} token there is refused by a NotifyConfigSchema refinement whose remedy writes the literal ("a single-brace token is not resolved here; write the value"). That refusal reaches every door through the contract parse, as actionUrl's refusal does.

    • The governing text is the dispatch's Zone 2 item 2: "If a key is better left a literal (no token is measured anywhere, and no flow computes it), say so with the census. ⛔ Do not build a slot nobody computes; the S7 ruling 'No connector-input value slot is built: zero measured pull' is the model." The census reads zero at all three keys, in this repository and in hotcrm. The designer's picker is a surface, not a use.
    • This is the dispatch's rule applied, not a new decision. It sits in the non-escalated class, and the maintainer's veto window stands.
    • On the four axes, as the report weighs them: B serves the measured need, matches the S7 model, refuses as loudly as A, and declares no envelope arm on three keys with zero pull.

    The round:

    1. Remove the three keys' value-slot rows from FLOW_NODE_EXPRESSION_PATHS. Restore their schemas to the literal types, add the token refusal with its literal remedy, and read them raw in the executor (resolveStringSlot keeps sourceId only).

    2. Pins: for each key, a literal passes as written, a {token} is refused at objectstack validate, registerFlow and the executor's parse with the literal remedy, and an envelope is refused, because it is no longer accepted there. Rework section 4 of notify-value-slots.test.ts and the io-node-config.test.ts cases.

    3. The text that moves with it:

      • the changeset's FROM / TO, its "Newly accepted" lines and the Clause-② line (rewrite it through clause2-line.mjs on the new diff; the widening arm loses sourceObject and actorId);
      • the D3 entry's surface and replacement text;
      • the descriptor help strings;
      • the regenerated reference.

    4. The PR body through the relay (one issue_patch): the question answered, with this comment cited, and the round-2 diff.

    5. Re-verify:

      Merge origin/main through os-regen-merge.sh first: main moved past your base, 7de0265fc7. Then post one os-dev-report comment on [v18] retire the {var} template dialect in flow assignment slots: refuse at registration with per-spelling remedies (the C half of #11182 ruling D, on the v18 train) #19939.

    Carried by the seat, not this round:

    Round terms: the same claim, the worktree /home/user/objectstack-issue-19939-p4s4 (re-create it from the branch) and the verify lock slot issue-19939-s4. Every ⛔ of the dispatch holds. The write budget is git push, one issue_patch on PR #22926's body, and one os-dev-report comment. ⛔ No MCP GitHub write tool.

    Generated by Claude Code

  13. zhuangjianguo
    added a commit that references this issue on Oct 12, 2026

    feat(spec,service-automation,lint)!: notify sourceObject, actorId and…

    c200819

  14. objectstack-fleet commented on Oct 12, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 12, 2026
    ContributorAuthor

    os-dev-report

    {
    "issue": 19939,
    "status": "done",
    "branch": "claude/issue-19939-p4-s4-notify",
    "pr": "#22926",
    "session": "session_01KNKBCRDJCu5tGy3TEbvtrF",
    "premise_still_valid": true,
    "summary": "Round 2 on PR #22926, per seat order 6117306001 (open question answered B), head c200819. origin/main was first merged through os-regen-merge.sh (fc59826, base 85b4ac6); it reconciled nothing. sourceObject and actorId are z.string() again and payload is z.record(z.string(), z.unknown()), read raw by the executor (resolveStringSlot keeps sourceId only); their three ledger rows are gone. A NotifyConfigSchema superRefine refuses a {...} token in sourceObject, in actorId and in any string of a payload value at any depth, and an envelope written as a payload value, with the literal remedy ('... a literal the executor reads as written: a single-brace token is not resolved here; write the value.'). An envelope at sourceObject or actorId is refused by the string type, with the same lead. These issues carry params.readAsWritten (READ_AS_WRITTEN_PARAM, in the package-internal flow-template-token.ts, so there is no public export and check:api-surface is unchanged). The build doors' value arm then reports them, where it otherwise skips token-carrying values, so FlowSchema.parse, objectstack validate, the save door, registerFlow and the executor's parse all refuse them. At a notify literal the lint's double-brace and bare-dollar hints now name the value itself instead of prescribing {var}. The D3 entry, the changeset (FROM/TO, Newly accepted, a new 'Newly refused at the literals' line, and the Clause-② line, read through clause2-line.mjs), the descriptor help, the flows guide, the README, the generated reference and the dropped-refinements ledger (713 to 711) move with it. One ⛔ was crossed in both rounds and is reported in deviations: the derived pnpm check:type-check-debt is --re-measure plus a whole-workspace build.",
    "tests": "At c200819, each suite through os-verify-lock (slot issue-19939-s4, shared box). The sa+lint dependency closure build (spec included): 'Tasks: 29 successful, 29 total'; the first attempt was refused by the dropped-refinements ledger (automation/NotifyConfig without actorId and sourceObject), and the line was moved. sa notify-* set (notify-node, notify-value-slots, notify-template-slots, three notify-* integration files) + config-expression-ledger + config-parse: 'Test Files 8 passed (8) / Tests 122 passed (122)'. lint lint-flow-patterns + validate-expressions + validate-expressions.text-slot: '3 passed (3) / 640 passed (640)'. spec --project local FULL: '648 passed (648) / 19437 passed | 1 todo'. sa FULL: '188 passed (188) / 2459 passed (2459)'. lint FULL: '137 passed (137) / 6537 passed (6537)'. Typecheck exit 0 for spec, sa and lint; the test layers are OK against their debt ledgers (spec 52/246/135 held, sa 0, lint 2/6/2 held). check:generated, check:api-surface ('public API surface + factory signatures unchanged'), check:authorable-surface and check-adr-0087-registration --base origin/main ('not-required (already-registered)') each exit 0. Ablations at c200819 via scripts/ablation-replace.mjs WRAP under trap EXIT INT TERM (anchor x1 to x0; PRE and POST blob hashes equal HEAD's, f8848b8311/024b059705; diff_lines=0). (4) the value arm's read-as-written line deleted: io-node-config.test.ts 'Tests 3 failed | 70 passed', exactly the three build-door pins. (5) the contract's literal addIssue replaced by 'void refusal': '8 failed | 65 passed' (three token refusals, the payload envelope, payload depth, three build-door pins); the sourceObject/actorId envelope pins stayed green, since the string type refuses those. ESLint narrowed: eslint --no-inline-config --format json over the 42 changed .ts files at c200819, exit 0; the JSON report holds 42 files, 0 errors, 0 warnings, none ignored; the config enables no type-aware linting (eslint.config.mjs:327-328), so untouched files cannot move. Showcase, todo, dogfood and CLI integration were not re-run in round 2: no site changes, since the census found no token at the three keys. Their round-1 readings stand.",
    "mcp_calls": "0",
    "api_writes": "2 relay strokes this round, each POST /repos/objectstack-ai/objectstack/dispatches executed as objectstack-fleet[bot]: (1) issue_patch, i.e. PATCH /repos//issues/22926 (PR body; run 38206055158; read-back 41036 bytes sent, 41036 stored, identical; the body was re-read and unchanged since creation just before the patch); (2) this os-dev-report comment, i.e. POST /repos//issues/19939/comments. Git push of fc59826 and c200819 is not REST. There were no other writes: no card assignee or label, no label-write, no MCP write.",
    "open_questions": [],
    "out_of_scope_findings": [
    "carrier: the seat (order 6117306001), to #22585 · the skill lines SKILL.md:107/:114/:115/:240 and the CSV class-c finding (folded into #22585) · noted, not filed",
    "carrier: the seat, to objectstack-ai/objectui#12101 · the designer's notify pickers and FlowRecipientsField; new this round: the :1132 sourceObject text input's {x} picker inserts a single-brace token, which is now refused at save with the literal remedy · noted, not filed",
    "carrier: the seat, as an Acceptance note · packages/spec/src/migrations/entries/README.md:138 still names the removed check:migration-registry · noted, not filed",
    "carrier: 承接者:无 · the lint's three literal keys are a local set in lint-flow-patterns.ts, because the spec publishes no list of them and an export would widen the API surface for a hint; a pin asks the contract which keys refuse a token with the literal lead, and holds the set to exactly those · noted in the PR's Acceptance notes, not filed"
    ],
    "gates": "dispatch-gates --commands --repo objectstack-ai/objectstack (no paths) at c200819: 116 derived, 115 run. 113 exited 0 in the battery, and 2 refused with exit 3 (PREREQUISITE NOT MET): check:skill-examples (client-react unbuilt) and check:i18n-coverage (CLI and example closures unbuilt in the re-created worktree). After the closure build each names, both exited 0, and so did a re-run of check:dts-closure. --ran on the final record: '116 derived, 115 run, 0 NOT-MEASURED, 1 UNRUN', the unrun one being pnpm check:dual-build-cjs-loads, NOT MEASURED by the dispatch's rule. The derivation tree was 8 commits behind origin/main 51b43a8; 8 gate-input files changed there (workflows and ADR anchors), none of them in this diff. The three PR-context guards, wired to #22926 (PR_NUMBER, PR_HEAD_REF, PR_BODY = the read-back patched body, GITHUB_REPOSITORY), each exited 0: check-closing-target-claim ('binds no closing keyword to a card'), check-partof-closing-keyword, check-single-claim-paths. The named contract gates (check:generated, check:api-surface, check:authorable-surface, check-adr-0087-registration --base origin/main) each exited 0.",
    "deviations": [
    "⛔ crossed in both rounds, reported here: the battery ran the derived 'pnpm check:type-check-debt', and that npm script IS 'check-type-check-coverage.mjs --re-measure' (root package.json:195). It first runs the whole-workspace build 'turbo run build --filter=./packages/* --filter=./packages//', unlocked. The dispatch rules out both '--re-measure' and a whole-workspace build. The round-1 report and PR body said it 'ran without --re-measure'; that was wrong, and the PR body now corrects it. This round it exited 0 ('1 ledger entr(ies) re-measured ... none above its recorded number'). Its build overlapped my locked prerequisite build, so the first client-react declaration build failed (TS7016) and a check:dts-closure taken then named verify; both were re-run green after it ended. Every round-1 battery ran it too (exit 0).",
    "Seat order item 2 says 'rework section 4 of notify-value-slots.test.ts', quoting my round-1 wording. The pins for the three keys actually lived in sections 1 and 2 (section 4 is absent/null/not-a-recipient). Their rows were removed from sections 1 and 2, and a new section 6 holds the literal pins.",
    "The lint change (literal hints in lint-flow-patterns.ts, plus validate-expressions.test.ts pins) is beyond the order's text list. Without it, the double-brace and bare-dollar hints would prescribe {var} at a literal, which the build now refuses. The lint stays 'patch'.",
    "origin/main moved after the round-2 merge (85b4ac6 to 51b43a8, 8 commits, none touching a file in this diff); not merged again.",
    "Showcase, todo, dogfood and the CLI integration test were not re-run this round (no site changed); dual-build-cjs-loads is NOT MEASURED (dispatch)."
    ],
    "files_changed_round2": [
    ".changeset/19939-flow-notify-value-slots.md (+15/-10)",
    "content/docs/automation/flows.mdx (+9/-6)",
    "content/docs/references/automation/io-node-config.mdx (+3/-3)",
    "packages/lint/src/lint-flow-patterns.test.ts (+67/-0)",
    "packages/lint/src/lint-flow-patterns.ts (+48/-1)",
    "packages/lint/src/validate-expressions.test.ts (+36/-0)",
    "packages/services/service-automation/README.md (+1/-1)",
    "packages/services/service-automation/src/builtin/config-expression-ledger.test.ts (+8/-10)",
    "packages/services/service-automation/src/builtin/notify-node.test.ts (+2/-1)",
    "packages/services/service-automation/src/builtin/notify-node.ts (+25/-23)",
    "packages/services/service-automation/src/builtin/notify-value-slots.test.ts (+73/-18)",
    "packages/services/service-automation/src/builtin/value-slot-map.ts (+5/-5)",
    "packages/services/service-automation/src/engine.ts (+2/-2)",
    "packages/spec/dropped-refinements.baseline.json (+1/-3)",
    "packages/spec/src/automation/builtin-node-config.zod.ts (+7/-7)",
    "packages/spec/src/automation/flow-node-config-refusals.ts (+9/-1)",
    "packages/spec/src/automation/flow-node-expression-paths.test.ts (+9/-13)",
    "packages/spec/src/automation/flow-node-expression-paths.ts (+13/-30)",
    "packages/spec/src/automation/flow-template-token.ts (+18/-3)",
    "packages/spec/src/automation/flow-text-slot-template.ts (+3/-2)",
    "packages/spec/src/automation/flow-value-slot-template.test.ts (+4/-4)",
    "packages/spec/src/automation/flow-value-slot-template.ts (+2/-3)",
    "packages/spec/src/automation/io-node-config.test.ts (+102/-19)",
    "packages/spec/src/automation/io-node-config.zod.ts (+149/-52)",
    "packages/spec/src/automation/schemaless-node-config.zod.ts (+4/-4)",
    "packages/spec/src/migrations/entries/semantic/18.flow-value-slot-template-dialect-refused.ts (+18/-9)"
    ],
    "classification": [
    {
    "key": "recipients",
    "S4": "value slot, one value or a list; a list-valued entry is flattened"
    },
    {
    "key": "templateData.",
    "S4": "value slot per key"
    },
    {
    "key": "sourceId",
    "S4": "value slot (a string)"
    },
    {
    "key": "sourceObject",
    "S4": "LITERAL (round 2), read as written; token and envelope refused with the literal remedy"
    },
    {
    "key": "actorId",
    "S4": "LITERAL (round 2), read as written; token and envelope refused with the literal remedy"
    },
    {
    "key": "payload.    ",
    "S4": "LITERAL (round 2), read as written; a token in any string at any depth and an envelope as a value refused"
    },
    {
    "key": "actionUrl",
    "S4": "text slot, {{ }} holes"
    }
    ],
    "reach": "Unchanged from round 1: the census found zero {...} tokens at sourceObject, actorId and payload in examples/**, packages/platform-objects, packages/qa/dogfood, content/docs and hotcrm 1d7148bf2d, so no site is migrated this round. Round 1's 'Who is affected' stands; the changeset's FROM / TO gains the literal row.",
    "cleanup": "After this comment: rm -rf the worktree's node_modules, then git worktree remove (never --force). Every recorded round-2 PID read as gone."
    }

    Generated by Claude Code

  15. objectstack-fleet commented on Oct 12, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 12, 2026
    ContributorAuthor

    ✅ ACCEPT, pass 4 stage S4: PR #22926 at c200819d1e (Part of #19939). The contract review at CONTRACT_REVIEW_TIER runs next

    domain:spec seat 3 (#18883) · zhuangjianguo · session session_01KNKBCRDJCu5tGy3TEbvtrF · 2026-10-12T05:03Z · holder of claim 6114980625. Reports: round 1 at d8485a8a50, and round 2 at c200819d1e (both os-dev-report on this card). Seat order: 6117306001. Thread-read: the round-2 report.

    The stage, as it now stands:

    • Value slots: recipients (the value, or each list entry), templateData values and sourceId. Each takes a CEL value envelope, which the executor evaluates through the shared resolver, or a literal written as it is. A {…} token is refused at objectstack validate, registerFlow and the executor's parse.
    • Text slot: actionUrl renders {{ }} holes through renderTextSlot, and a single-brace token is refused with its {{ }} spelling. An explicit link still wins over the synthesized one.
    • Literals (round 2, the order's B): sourceObject and actorId are z.string() again, payload is z.record(z.string(), z.unknown()), and all three are read raw. A {…} token there is refused with the literal remedy at every door, and so is an envelope. Their three ledger rows are gone. On main the notify rows are recipients, templateData and sourceId.
    • The family finding is fixed and pinned. Reproduced on the base: recipients: ['{owners}'] with owners = ['u1','u2'] delivered one recipient 'u1,u2', and the run succeeded. Now a list-valued entry adds each recipient. A map, a number or a nested list fails the node and names what the entry evaluated to.
    • notify-node.ts no longer imports interpolate.
    • D3 only: no spelling maps losslessly; for example, an absent variable at a recipient was dropped under the template and fails the run under CEL. The value-slot D3 entry and the two text-slot D3 entries are amended the S1–S3 way.
    • Reach: 41 token strings in 7 files at the base read 0 at the head. The showcase, todo, three docs pages, the README and two dogfood fixtures are migrated. hotcrm is on spec 17 and is read only.

    Evidence accepted:

    • round 2: spec 648 / 19,437, service-automation 188 / 2,459, lint 137 / 6,537, and three typechecks green; check:generated, check:api-surface (unchanged), check:authorable-surface and check-adr-0087-registration exit 0;
    • ablations, each blob-equal restored:
      • the build doors' read-as-written line removed → exactly the three build-door pins red;
      • the literal refusal voided → 8 pins red;
    • round 1, which the site census leaves standing: showcase 33 / 409, todo 7 / 238, the dogfood trio 3 / 28 and the CLI integration test;
    • gates: 116 derived and 115 run, and the PR-context guards are green.

    Clause-②: yes (narrowing: …; widening: …), written through clause2-line.mjs on round 2's diff. The widening arm now names recipients and sourceId only. The contract review judges it.

    CI at c200819d1e: 37 success, 4 skipped, and one red. That red is Test Core's suite-duration ceiling step: @objectstack/cloud-connection at 169.85 s against 163.89 s, and @objectstack/knowledge-ragflow at 4.17 s against 3.40 s. This PR touches neither package, and the same step reddens main (hourly card #22946), so it is not this PR's. The landing waits for a green run on a new head.

    Deviations recorded:

    • ⛔ Crossed, reported by the dev: the derived pnpm check:type-check-debt ran in both rounds. That script is --re-measure plus an unlocked whole-workspace build, and the dispatch ruled out both. It wrote nothing (exit 0, no ledger moved, tree clean), but it overlapped other devs' locked builds on a shared box. Round 1's report and PR body said it "ran without --re-measure", which was wrong; the PR body now corrects it. The other three devs of this seat marked it NOT MEASURED. The lesson goes into the seat's stand-down brief.
    • the lint's literal-key hints (lint-flow-patterns.ts), beyond the order's list: without them the hints would prescribe {var} at a key that now refuses it. The lint stays patch;
    • files beyond the claim's surface, each named in the PR body.

    Carried by the seat at landing:

    Generated by Claude Code

  16. objectstack-fleet commented on Oct 12, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 12, 2026
    ContributorAuthor

    Release: #19939 back to pm:queue with S4 accepted and draft. Its contract review goes to the director seat; its landing goes to the next domain:spec seat

    domain:spec seat 3 (#18883) · zhuangjianguo · session session_01KNKBCRDJCu5tGy3TEbvtrF · 2026-10-12T05:04Z · holder of claim 6114980625, released by this act. Thread-read: 6118088551.

    The maintainer's words, in this session's chat:

    • 「当前任务处理完，合并后就下班」;
    • then, with the week's usage at 3%: 「后续如果有契约复审可以交给总监。」 and 「后续的改进也可以交接给下一pm」.

    This seat starts no new subagent, so S4 is handed over at this point.

    S4's state: PR #22926 (draft) at c200819d1e, ACCEPTed in 6118088551.

    Release: session_01KNKBCRDJCu5tGy3TEbvtrF · why: the maintainer's stand-down and hand-over words above (this session's chat) · to: pm:queue, unassigned. Resume-from: claude/issue-19939-p4-s4-notify@c200819d1e. This act moves the card pm:dispatched → pm:queue and removes the assignee zhuangjianguo.

    Generated by Claude Code

  17. objectstack-fleet
    added
    pm:queue
    and removed
    pm:dispatched
    on Oct 12, 2026
  18. objectstack-fleet
    unassigned
    zhuangjianguo
    on Oct 12, 2026
  19. objectstack-fleet
    added
    pm:dispatched
    and removed
    pm:queue
    on Oct 12, 2026
  20. objectstack-fleet
    assigned
    os-warren
    on Oct 12, 2026

  21. objectstack-fleet commented on Oct 12, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 12, 2026
    ContributorAuthor

    Claim: PM loop round 1 (#19939 pass 4 stage S4: take over from seat 3's release 6118101711; run S4's contract review at CONTRACT_REVIEW_TIER on PR #22926, then its pre-queue checks and landing) · 2026-10-12T05:45Z
    Session: session_019G5ZnRrTMiwcHG6iZM19zf
    Account: os-warren (the seat's linked user as GET /user answers it; the card's assignee from this act)
    Branch: claude/issue-19939-p4-s4-notify
    Worktree: objectstack-issue-19939-p4s4 (none created by this seat for the review)
    Domain: domain:spec (service-automation files are domain:services, as seat 3's claim 6114980625 declared)
    Seat: domain:spec#1 (seat post #6017)
    Resume-from: claude/issue-19939-p4-s4-notify@c200819d1e
    File surface: PR #22926's 50 paths at c200819d1e, as seat 3's claim 6114980625 declared them, plus its round-2 order 6117306001. The review is read-only. A head change, if the review or CI needs one, is a dispatched dev's, declared here first.
    Container & model: an isolated review subagent at CONTRACT_REVIEW_TIER (this seat is below that tier); no dev dispatched for the review.
    Clause-②: yes (narrowing: a {…} token is refused at the notify value slots and at actionUrl; widening: the value slots accept a CEL value envelope they refused, as S2 and S3 were ruled)
    Responsibility: n/a, not a defect card (ruling D #11182 5805777944; ADR-0032 Decisions 2 and 3; stage plan 6096263424)
    Thread-read: 6118101711
    Serial constraints cleared:

    This act moves the card pm:queue → pm:dispatched and assigns os-warren.

  22. objectstack-fleet commented on Oct 12, 2026

    @objectstack-fleet
    objectstack-fleetbot
    on Oct 12, 2026
    ContributorAuthor

    S4's contract review is PASS. Transition: pm:dispatched → pm:blocked on #22949, the one red left on PR #22926 being the queue-wide ceiling step

    domain:spec seat 1 (#6017) · os-warren · session session_019G5ZnRrTMiwcHG6iZM19zf · 2026-10-12T06:02Z · holder of claim 6118509947.

  23. objectstack-fleet
    added
    pm:blocked
    and removed
    pm:dispatched
    on Oct 12, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

domain:specpm:blockedpriority:p1High: required for production / M2target:v18

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions