fix(scripts): spell pnpm flags so a global pnpm 11+/12 frontend parses them (--config.production=false, --silent)

[objectstack-fleet]

Problem

A contributor whose globally installed pnpm is a newer @pnpm/exe frontend (11.x / 12.x) cannot get pnpm dev past check:console-sha once the objectui pin moves. The repo pins packageManager: pnpm@10.31.0, and pnpm --version dutifully reports 10.31.0 — but the new frontend parses the argv itself (clap-style) before it hands off to the pinned version, and it rejects two spellings this repo's scripts use:

spelling	where	new frontend says
pnpm install … --prod=false	scripts/build-console.sh (pnpm objectui:build)	error: unexpected value 'false' for '--prod' found; no more were expected — exit 2, nothing installed
pnpm -s SCRIPT	packages/spec/scripts/check-generated.ts (every check:generated gate), scripts/check-regen-pending.mjs (the pre-commit regen gate), ownerRunCommand({ silent })	error: unexpected argument '-s' found — exit 2, which those callers read as "stale"

CI never sees it: Corepack runs the real pnpm 10.31 binary directly, which accepts all of these.

Change

• build-console.sh: --prod=false → --config.production=false.
• Every spawned pnpm -s → pnpm --silent (the long form of the same flag): check-generated.ts, check-regen-pending.mjs, regen-artifacts.mjs ownerRunCommand, and the self-test pin for it in git-merge-regen.mjs. Comments that describe the spawn follow; two comments that record a historical measurement taken with pnpm -s are left as written.
• dispatch-gates.mjs names pnpm -s ${script} as a live specimen of the shell-form class; the specimen and its self-test fixture string now read --silent.

No workflow change: every .github/workflows/** call already spells --silent.

Evidence

All probes in a scratch workspace pinned to packageManager: pnpm@10.31.0, run with both binaries: OLD = the real pnpm 10.31.0 @pnpm/exe binary (what CI runs), NEW = the global @pnpm/exe 12.3.4 frontend.

Install: which spelling parses on both AND still installs devDependencies. One dependency + one devDependency; dev= is whether the devDependency landed. First, the controls — what actually turns production mode on under pnpm 10.31 (measured, OLD, no flag):

context	dev
nothing	yes
NODE_ENV=production	yes — pnpm 10 does not skip devDependencies on NODE_ENV alone
npm_config_production=true	no
.npmrc production=true	no
--prod	no

Then each candidate, against the two contexts that do skip devDependencies (identical rows for both contexts, so shown once):

spelling	OLD	NEW
(none)	rc 0, dev no	rc 0, dev no
--prod=false	rc 0, dev yes	rc 2, unexpected value
--production=false	rc 0, dev yes	rc 2, unexpected value
--no-production	rc 0, dev yes	rc 2, unexpected argument
--no-prod	rc 0, dev yes	rc 0, dev yes
--config.production=false	rc 0, dev yes	rc 0, dev yes

--config.production=false is chosen over --no-prod because the new frontend's own help documents --config.KEY=VALUE as a universal rc-option, and it names the exact setting being overridden.

Run: --silent is byte-identical to -s. For a root script and a --filtered one, each both passing and failing (exit 3 / 4): stdout, stderr and exit code of OLD -s compared with cmp against OLD --silent and NEW --silent — 8/8 identical (--reporter=silent also 8/8). Control: -s vs no flag differs, so the comparison is live. --loglevel=silent is NOT equivalent (still prints the > pkg script header).

Sweep: nothing else the repo spells is rejected. Every other option shape the repo's scripts pass to pnpm (--filter/--filter=, -r list --depth -1 --json, exec … -e, trailing script args such as --maxWorkers=2 --update -e x, run x -- --y, install --no-frozen-lockfile / --offline, pack --pack-destination … --silent, store path --silent, -v) parses identically under OLD and NEW.

Repo, under the global 12.3.4 shim, no PATH workaround (command -v pnpm → ~/Library/pnpm/bin/pnpm):

command	before (main 0326b32952)	after (this branch)
pnpm objectui:build	exit 2 at the objectui install	rc 0 — install ran (Packages: +1566, devDependencies included), console dist built; pnpm check:console-sha then passes, so pnpm dev gets past it
pnpm --filter @objectstack/spec check:generated	rc 1, 14/14 gates red, each unexpected argument '-s'	rc 0, 14/14 up to date
node scripts/check-regen-pending.mjs --self-test	rc 1 (the stub-gate cases red)	rc 0
node scripts/git-merge-regen.mjs --self-test	—	rc 0
node scripts/pm/dispatch-gates.mjs --self-test	—	rc 0, 2144 cases

The "before" column is the same command with only the touched file restored from main into the working tree, after the fix was committed.

Repo, under the real pnpm 10.31.0 binary first on PATH (what CI runs): check:generated rc 0 (14/14), pnpm check:merge-driver rc 0, pnpm objectui:build rc 0. (In that run the objectui install was a no-op over the already-populated cache tree; the install semantics under 10.31 are the scratch table above.)

eslint over the six changed JS/TS files: clean. bash -n scripts/build-console.sh: clean.

Notes

• No changeset: only scripts/** and packages/spec/scripts/** change, and neither ships in any published package (@objectstack/spec's files does not include scripts/) — skip-changeset.

---

Generated by Claude Code

[github-actions]

📓 Docs Drift Check

Nothing in this diff resolved to a documentable surface (no symbol, route or SDK anchor derived from 0 changed package(s)), so this run has no opinion about the docs.

What this run could not see

• a page that states a rule by its inputs shares no identifier with the emitter that implements the rule, so an emitter-only diff cannot list it — not on this run and not on any run. Measured on fix(driver-sql): emit varchar(maxLength) for a text field a declared index keys on #11430: content/docs/protocol/objectql/types.mdx documents the text-family column mapping by the ObjectQL type names it maps FROM (text / textarea / html) while the diff changed createColumn; it went unlisted, and it was the page that diff falsified, in four places. No shared token exists to detect this on, so a rule your change carries has to be re-read by hand in the pages that restate it.
• a key NAME is not a key, so the hand re-read the line above prescribes can land on the wrong schema. The same spelling is authorable on one governed type and a [REMOVED] tombstone on another for each of active, aria, joins, objects, template, tools and version (censused on [finding] tools is a key on BOTH AgentSchema (tombstoned, dead) and SkillSchema (live, cloud-attested), so a name-based search attributes skill examples to the agent key — it produced a false stop-the-line alarm on PR #19059 #19093 over the liveness ledger's governed types, top-level keys); nothing in a search result distinguishes the two, so a grep hit on a LIVE example reads as evidence about the DEAD key. Measured on fix(spec): the agent.tools liveness row says dead — it claimed live on a key the schema tombstoned #19059: content/docs/ai/agents.mdx was reported as contradicting the agent.tools tombstone over its tools: example at :161, which is inside the defineSkill({ block opened at :155 — the page was already correct. Settle ownership by PARSING the value against both schemas, never by the name: that literal PASSES SkillSchema, and as an AgentSchema it FAILS at tools with the tombstone prescription. ⛔ These names are not the whole class — a key retired through a .strict() guidance map leaves no tombstone in the walked shape and none of them here (tool.category, live as AIToolDefinition.category).

Coarse fallback — 0 page(s) merely mention a changed package (the pre-#9192 predicate, kept for the deliberately-wide backstop): node scripts/docs-audit/affected-docs.mjs --json 4fe8b3def02ff9803a52c2fb8b97778d5bf45597 → packageMentionDocs.