@lotics/cli 0.258.0 → 0.259.0

package/dist/src/client.d.ts

@@ -1608,6 +1608,18 @@ export declare class LoticsClient {
             thumbnail_url?: string;
         };
     }>;
+    /** A file of the app under another name — a new file over the same bytes, for a save to put in the old one's place. */
+    appRenameFile(app_id: string, file_id: string, filename: string): Promise<{
+        file: {
+            id: string;
+            filename: string;
+            mime_type: string;
+            url?: string;
+            thumbnail_url?: string;
+            size?: number | null;
+            created_at?: string;
+        };
+    }>;
     appGetRecordComments(app_id: string, record_id: string): Promise<unknown[]>;
     appCreateRecordComment(app_id: string, record_id: string, body: {
         content: string;

package/dist/src/client.js

@@ -1100,6 +1100,14 @@ var LoticsClient = class {
       body
     );
   }
+  /** A file of the app under another name — a new file over the same bytes, for a save to put in the old one's place. */
+  async appRenameFile(app_id, file_id, filename) {
+    return this.request(
+      "POST",
+      `/v1/apps/${encodeURIComponent(app_id)}/files/${encodeURIComponent(file_id)}/rename`,
+      { filename }
+    );
+  }
   // App-scoped record comments — the `lotics app dev` loop forwards the iframe's
   // `comments.*` ops to these (production routes them through the iframe host).
   // App authority + tenant floor are enforced server-side; these are thin.

package/docs/cli_reference.md

@@ -54,7 +54,7 @@ Per-command syntax, flags, contracts, and gotchas for the public `lotics` CLI. S
 | `lotics app codegen [path]` | Regenerate `.lotics/*` from the manifest + workspace schema **without a deploy**. The three `.d.ts` companions (`app_{workflows,queries,agents}.d.ts`) are always rewritten (synchronous, no network). When credentials resolve, also rewrites the **runtime** `.lotics/app_fields.ts` — a real `.ts` exporting `F` (table→field→`"fld_…"`), `OPT` (table→select-field→option→`"opt_…"`), `TBL` (table→`"tbl_…"`) and `GRP` (member group→`"grp_…"`) keyed by display-name aliases, for every table the app's queries READ plus every table its bound workflows WRITE (each binding's recorded `table_ids`, so a table no screen reads is still addressable by alias and a rename fails `tsc` instead of the body), plus every member group in the organization — which one a screen names is not knowable from the manifest, and a `GRP` narrowed by a failed read is a map that is wrong rather than absent, so a failure writes nothing. **There is one form, and that is what makes a starter's source portable**: the keys are slugified DISPLAY NAMES and a starter carries its labels verbatim, so running codegen in a copy's own workspace emits the same keys pointing at that workspace's ids — no binding fetched at load, no prebuilt bundle to keep in step. Beside it, `.lotics/app_siblings.ts` exports `APP` — every app the workspace's model declares, by the model's alias for it (`APP.<alias>` → `"app_…"`), read off the workspace's model binding — so a hand-written app opens a sibling as `openApp(APP.<alias>, route)` and holds no `app_` id; an app archived here is absent, so a call naming it fails `tsc`. The binding is an admin's read: a key that cannot read it keeps the file on disk and says so. Also refreshes each bound workflow's `.lotics/workflows/<alias>.globals.d.ts` + re-wraps its EXISTING `src/workflows/<alias>.ts` body in the current envelope (strips + re-wraps; never re-fetches the body, so local edits survive). **`.lotics/` is reconciled to the manifest, not merely added to** — a `<alias>.globals.d.ts` whose alias the manifest no longer declares is DELETED. Only that exact filename shape is removed; anything else in the directory is left alone. The reconcile runs before the credential branch, so it happens offline too. The authored counterpart is never deleted — a `src/workflows/<alias>.ts` the manifest does not declare is NAMED instead (`check` and `set` both take their alias set from the manifest, so editing an undeclared body is a silent no-op). A getApp / binding / schema / dts-fetch failure is non-fatal (warns, keeps the last-generated files). **Re-silvers `package.json#lotics.agents`** from the live app row whenever its `inputs`/`outputs` disagree, then rewrites the agent `.d.ts` from the refreshed block: that block is a mirror AND the offline seed for `useAgentRun` typings, so a stale copy types the app against an agent that does not exist. The write is surgical and order-preserving, so it changes only the fields that actually differ. A hand edit to that block is therefore reverted — it never changed the agent anyway; to change one, `set_app_agent`. **Types are written for every DECLARED alias, body file or not** — the dts is rendered from the declaration, which is the whole point of the declare → codegen → write → check loop — and codegen NAMES each alias it wrote types for without a body, with the next step. It used to return early on a missing or blank `src/workflows/<alias>.ts` and say nothing, which left `workflow check` pointing at `app workflow pull` (which writes nothing for an alias the server has never bound) on one path and at `app codegen` — the command that had just declined to write them — on the other. **A body whose helper sits ABOVE the `__workflow` wrapper is refused by name and line** rather than wrapped a second time: the strip peels a wrapper only when it is the first line after the header, so a top-level declaration between the two used to leave the whole file read as the body and the re-wrap nested an envelope per run. Move the helper inside the wrapper — the body is one expression sequence. |
 | `lotics app kit <path-to-packages/app-runtime>` \| `lotics app kit --published` | **The app's ONE range, `@lotics/app-runtime`, from a CHECKOUT, or back to the range it listed before**, so a runtime change is proven on a real app before it is published. Run inside the app directory; the path names `packages/app-runtime` in your clone (a kit change reaches an app through the runtime that depends on it — `LOTICS_UI_SRC` links the kit's source into `app dev`). It runs the runtime's own `npm run build`, `npm pack`s it into the app's `.lotics/kit/` (replacing any earlier pack, so the directory holds one tarball), removes the installed copy and the lockfile's entry for it, installs by file specifier, and then **proves the install**: the first non-dot file of the directory the package's own `package.json#files` publishes is hashed on both sides, and a mismatch — or an absence — is a refusal naming the file. That proof is the point. A repack under the SAME name is the normal case (a working copy's version does not move between builds) and the lockfile pins the first tarball's integrity, so on some npm versions the install is a no-op that reports success while the app keeps building against an hour-old kit; nothing else in the loop can tell. **Either way the rest of the manifest follows the runtime**: `@lotics/ui`, where the app lists it, moves to the range the runtime depends on — two ranges are two copies of the kit — and an app still on `@lotics/app-sdk` (the package the SDK was before it moved into the runtime) is MIGRATED: every `@lotics/app-sdk` and `@lotics/app-sdk/router` specifier under `src/` and in the Vite config becomes `@lotics/app-runtime/sdk` and `@lotics/app-runtime/router`, the dependency is dropped, and the summary names each file rewritten. **Refusals, all before anything is written**: a directory that is not a Lotics app (no `package.json#lotics.workspace_id`); a path that is not `@lotics/app-runtime`; a runtime that declares no `@lotics/ui` range while the app lists the kit; `--published` given a path; a build that failed, named with npm's own output tail. **It records what it did in `package.json#lotics.kit.@lotics/app-runtime`** (`version`, `tarball`, the `source` checkout, `packed_at`), which is what `lotics app check` warns about and what `lotics app deploy` REFUSES on unless you pass `--allow-local-kit` — `.lotics/` is excluded from the source archive a deploy uploads, so an app whose manifest names `file:.lotics/kit/…` ships a bundle that runs and a source tree the next clone cannot install. The record also keeps what the app listed before the FIRST checkout went in (`replaced` — a repack keeps it), and `--published` undoes the trial: the runtime and the kit go back to exactly those specs, so a kit trial never moves an app to another line on its way out; an app with no such record (never given a checkout, or given one by a CLI that recorded none, which the run says) moves to `^<latest>` from the registry and the kit to the range that published version depends on. Either way the tarball and the record are deleted, the lock entry and the installed copy are cleared, and a plain install resolves the published package. Needs no credential — npm, the filesystem and (for `--published` with no prior range) the registry. |
 | `lotics setup <apg_id \| model.json> [--email <addr>] [--json]` | **The whole first run, in one command.** Creates an account when this machine has no credential (the same call `auth signup` makes — `--name` and `--timezone` apply), then fills its workspace, then prints the one-time sign-in link. **When that email already has an account it hands over to the `lotics auth login` flow** — it prints the sign-in page to open and the code it must show, and **exits 1 having created nothing**; the person presses Confirm and runs the same command again, which collects the key and carries on into the copy or the model. (`--wait` holds the terminal through the Confirm instead, finishing in one command.) The re-run is not refused for naming an `--email` it is now signed in as — that address IS the account it holds, not a second one. **The argument decides which of the two forms this is, by SHAPE**: a `*.json` file is a workspace MODEL — in either of ITS two forms, spelled out or `{"from": "<preset-slug>", …}` — and anything else is a package id copied through `library init`. The suffix decides it alone — asking the filesystem would answer a long library id with `ENAMETOOLONG` instead of with a verdict — and a model is checked OFFLINE before an account is created, because a file with a typo in it must not leave an organization behind. The model form creates no apps, so its sign-in link lands on the first table it made. It sends no `adopt`: an entity whose `label` already names a table in the workspace is REFUSED with every collision named, and the refusal adds the line the server cannot — `lotics scaffold apply <model.json>`, the verb that adds to the workspace you already have. It exists because the two-command form has a seam where the FIRST command exists only to produce a credential for the second, and a caller pasting a prompt has to get both right. **`--email` is only for creating an account**: with a credential already resolvable it is REFUSED rather than obeyed, because the two can name different organizations and preferring either one silently copies a package into an org the caller did not name — the message says how to do each thing on purpose. Without it, `setup` copies into the account you already have and is a pure alias for `library init`. A path positional is accepted and IGNORED with a warning — `setup` writes nothing to disk — so a prompt that passes one still runs. **`--json` prints one object on stdout and nothing else** — `organization_id`, `workspace_id`, `app_ids` (alias → id), `apps` (each app's `version_number`, or its `error`), `signin_url`, and `created` — which NAMES what landed (`tables`, `templates` and `knowledge_docs` are alias arrays; `sample_records` is a row count, since rows are not named things). Aliases rather than counts because the next question is about a particular artifact: a copied template carries the publisher's wording and a copied knowledge doc describes how they work, so "which of these should be mine?" is the conversation a copy starts, and a count cannot begin it. **The model form emits `entities`, `roles`, `record_ids` and `rows_skipped`** in place of `app_ids` / `apps` / `created` — a model creates no apps and nothing named for a copier to review. **The model form also runs the file's `apply` list** — each named package copied in after the tables exist, with that entry's `bind`, in order, stopping at a refusal with everything before it kept — and emits `applied: [{package, apps}]` beside them; **the sign-in link then lands on the FIRST app any applied package created**, falling back to the first table when the model applied none. Plus a `warnings` array carrying everything the prose form would have said out of band — an unbindable knowledge doc, a sign-in link that could not be minted, the publisher's-code disclosure, an app that landed without a version. A warning is never merely silenced: when the command fails with an error before it can emit, the ones it had collected go to stderr alongside it. Reachable with no install: `npx -y @lotics/cli setup …`. |
-| `lotics scaffold check <model.json> [--json]` | **Prove a model before anyone sees it — no network, no credential**, unless the file names a preset. ONE parse of the whole file against the model schema (strict, so `tabels` or `row` is an error rather than a silently dropped key, and a model cannot express what only a starter bundle carries: `fixtures`, `knowledge`, `knowledge_expects`, a file-backed `excel`/`word`/`pdf-form` template; its `apps` are stated over the model's entities, never built code; a retired key — `field_roles`, `table_workflows`, `connections` — is refused by name with what replaced it), then `validateWorkspaceModel` — the caps, every cross-reference, and the first rows themselves (a field the entity does not declare, an unknown option alias, a link naming no row in the file, a duplicate `ref`, a date that is not one, a value on a platform-computed field, a files cell that is not a relative path beside the model or a `fil_` id, a document path with no file beside the model), then `write_rules`, then `records` and `apps`: an entity an app lists, opens or picks with no `records` entry, a title that is an autonumber or a code, the `record.facts` or `record.blocks` that `sections` replaced, a section holding nothing, a field, a block or an act placed twice on one record, an `at` on a record with no status, an act at a staged section's foot offered while the section is not the work, a `where` on a field that is not a single select, a block whose entity does not link to the record, a `when` on a field that is not a single select, an act setting a field or option that does not exist, a check on a field that is not a yes/no or text formula, a register or rows-block entity with no `singular`. **Reports EVERY problem in one run**, each as `<path>: <message>` in the file's own keys (`entities.0.fields.1.type`, `apps.0.acts.1.set.stage`), so fixing a model is not a round trip per mistake. Exits 1 when there is one; exits 0 with the counts (`N tables, N fields, N links, N views, N roles, N rows`, plus `N apps` when the file states any), then **each app as its reader will see it** — the register's row, columns, filters, add and order; the record's door, header and the buttons beside its ⋯, checks, its sections as the reader scrolls them — the stage each is the work of, its fields, its blocks and the acts at its foot — its thread, and its acts with every `when`, `requires` and blocking check — with every value a default supplied marked `(default)`, and a line naming `lotics app preview <model.json>#<app>`, all on stdout. `--json` replaces both with one object and nothing else: `{ok: true, …counts, resolved: {<app>: …}, notes}` or `{ok: false, findings: [{path, message}]}`. **A `preset` is checked as N models, not one** — every variant merged onto the base (its added entities, and its added fields keyed by entity) and put through the same rules, each finding addressed `preset.variants.<slug>.<path>`, so a preset ships with every branch proven: the branch nobody took is the one that fails in the workspace of whoever takes it, who is the one reader who cannot fix it. A variant's `fields` key naming no declared entity is a finding too — the merge keys on the entity, so a typo'd alias adds those fields to nothing. Same verdict the server reaches, because it runs the server's own functions out of `@lotics/shared` rather than a second implementation of them. **A file written as `{"from": "<preset-slug>", …}` is resolved first** — one GET of that preset's file on the website — and that read is the one step on this path that needs the network; it says so when it cannot make it, and a slug nothing serves is answered with the slugs there ARE, read from the listing, rather than with a 404 the author cannot spell their way out of. Resolution is pure (`resolveModelFrom` in `@lotics/shared`): the named variants merged onto the preset's base in order, then `rename` through the same `applyBinding` a `--bind` goes through, then the file's own `entities` appended, and the preset's `records` and `write_rules` merged with the file's, entity by entity (the file's entry replacing the preset's). What comes out is the full form and goes through everything above unchanged, so a `from` file cannot reach a workspace by a route the full form does not. A variant slug the preset does not declare, an alias `rename` names that it does not declare, and a renamed label that is already another table's are each a finding rather than a silent drop — a branch quietly ignored scaffolds the base and looks like it worked. |
+| `lotics scaffold check <model.json> [--json]` | **Prove a model before anyone sees it — no network, no credential**, unless the file names a preset. ONE parse of the whole file against the model schema (strict, so `tabels` or `row` is an error rather than a silently dropped key, and a model cannot express what only a starter bundle carries: `fixtures`, `knowledge`, `knowledge_expects`, a file-backed `excel`/`word`/`pdf-form` template; its `apps` are stated over the model's entities, never built code; a retired key — `field_roles`, `table_workflows`, `connections` — is refused by name with what replaced it), then `validateWorkspaceModel` — the caps, every cross-reference, and the first rows themselves (a field the entity does not declare, an unknown option alias, a link naming no row in the file, a duplicate `ref`, a date that is not one, a value on a platform-computed field, a files cell that is not a relative path beside the model or a `fil_` id, a document path with no file beside the model), then `write_rules`, then `records` and `apps`: an entity an app lists, opens or picks with no `records` entry, a title that is an autonumber or a code, the `record.facts` or `record.blocks` that `sections` replaced, a section holding nothing, a field, a block or an act placed twice on one record, an `at` on a record with no status, an act at a staged section's foot offered while the section is not the work, a `where` on a field that is not a single select, a block whose entity does not link to the record, a `when` on a field that is not a single select, an act setting a field or option that does not exist, a check on a field that is not a yes/no or text formula, a register or rows-block entity with no `singular`. **Reports EVERY problem in one run**, each as `<path>: <message>` in the file's own keys (`entities.0.fields.1.type`, `apps.0.acts.1.set.stage`), so fixing a model is not a round trip per mistake — one of the model's own rules ending with the rule's id (`[register.filter]`), and each rule named printed once more under the list in the one sentence `lotics docs model/<rule>` states it in. Exits 1 when there is one; exits 0 with the counts (`N tables, N fields, N links, N views, N roles, N rows`, plus `N apps` when the file states any), then **each app as its reader will see it** — the register's row, columns, filters, add and order; the record's door, header and the buttons beside its ⋯, checks, its sections as the reader scrolls them — the stage each is the work of, its fields, its blocks and the acts at its foot — its thread, and its acts with every `when`, `requires` and blocking check — with every value a default supplied marked `(default)`, and a line naming `lotics app preview <model.json>#<app>`, all on stdout. `--json` replaces both with one object and nothing else: `{ok: true, …counts, resolved: {<app>: …}, notes}` or `{ok: false, findings: [{path, message, severity, rule}]}` (`rule` on a finding of the model's own rules). **A `preset` is checked as N models, not one** — every variant merged onto the base (its added entities, and its added fields keyed by entity) and put through the same rules, each finding addressed `preset.variants.<slug>.<path>`, so a preset ships with every branch proven: the branch nobody took is the one that fails in the workspace of whoever takes it, who is the one reader who cannot fix it. A variant's `fields` key naming no declared entity is a finding too — the merge keys on the entity, so a typo'd alias adds those fields to nothing. Same verdict the server reaches, because it runs the server's own functions out of `@lotics/shared` rather than a second implementation of them. **A file written as `{"from": "<preset-slug>", …}` is resolved first** — one GET of that preset's file on the website — and that read is the one step on this path that needs the network; it says so when it cannot make it, and a slug nothing serves is answered with the slugs there ARE, read from the listing, rather than with a 404 the author cannot spell their way out of. Resolution is pure (`resolveModelFrom` in `@lotics/shared`): the named variants merged onto the preset's base in order, then `rename` through the same `applyBinding` a `--bind` goes through, then the file's own `entities` appended, and the preset's `records` and `write_rules` merged with the file's, entity by entity (the file's entry replacing the preset's). What comes out is the full form and goes through everything above unchanged, so a `from` file cannot reach a workspace by a route the full form does not. A variant slug the preset does not declare, an alias `rename` names that it does not declare, and a renamed label that is already another table's are each a finding rather than a silent drop — a branch quietly ignored scaffolds the base and looks like it worked. |
 | `lotics scaffold apply <model.json> [--json]` | **Create the model in this workspace**: its tables, fields, select options, links, views, roles and first rows, through `POST /v1/workspaces/scaffold`. Runs `check` first, so a bad file never reaches the network, then resolves and ANNOUNCES its workspace (`lotics → <org> / <workspace>` on stderr) before writing — it is a destructive path. **Additive and re-runnable**: it is the verb that sends `adopt`, so an entity whose `label` already names a table here BINDS to that table and gains the fields, options and views it is missing, while `setup` refuses that same label. No stored value is ever changed or deleted. A field the workspace already had takes the model's `default`, the `format` of its text or number and a number's `unit` — how a new row starts and how its values read — and keeps everything else; a date's format is left, since changing it would rewrite every stored date, and so is a unit moved between two of one dimension (`kg` to `t`), since it would convert every stored figure — `scaffold diff` names both as what apply leaves. So applying the same model twice changes nothing the second time — and a declared PAIRING (`sync_both_ways` / `paired_field_alias`) over a link this workspace already has one-way is REFUSED before the first write, naming the alias and `update_fields sync_both_ways`, because a pairing is only ever created with the link and adoption would otherwise finish clean over a half-paired link. **After the first run the WORKSPACE remembers what each alias became**, so every later run binds entity, field, select option, template and role BY ID and a relabel on either side is a RENAME rather than a second table: the run reports each moved name with the command that reconciles it (`lotics field rename` moves the platform, the file and every bound app together; `update_table` moves a table's name), and applies anyway — which of the two names is right is the author's to decide, and the run bound the thing the model has always meant either way. A bound target the workspace no longer holds is refused by name, with `restore_table` and `lotics scaffold unbind` as the two ways out. **Rows land only where every bound table is empty**: one bound table already holding records and none are written anywhere, because sample rows landing among a customer's real ones cannot be told apart from them — it says so and reports `rows_skipped`. Prints `created`/`adopted` per entity with its table id **and the delta that landed on it** (`+8 fields, +2 options, +1 view, 2 fields set to the model's default or format`, and nothing where the run changed nothing) — `adopted` alone cannot report the columns, options and views a later version of a model puts on a table that is already the owner's, and the only other proof was a full re-export and a diff — marking `(bound by name)` the one case where a NAME decided which table the model points at; the same `created`/`adopted` per ROLE with its group id — an adopted role binds a group that already exists, which is how one silently inherits another workspace's members — and rows written per entity. **The binding is the workspace's, not the file's**: a model file is never committed, so memory kept beside it is one author's disk — absent for a teammate, on a second machine or after a delete, and each of those falls silently back to the label join that grows the twin. It holds what each entity, field, option, role and template alias became (`tbl_`/`fld_`/`opt_`/`grp_`/`tpl_`), plus one entry per first ROW under its own `<entity>:<ref>` and `documents` (path → `fil_`), and is read by `diff`, `--documents`, `app create --from`, `app regenerate --from` and `workspace build` before any of them reads a label. **Then it copies in every package the file's `apply` list names, in order** — each one a `library init` with that entry's `bind` and `no_sample_data`, and each sending `adopt`, because by then the workspace holds exactly the tables this same run just created. Order is load-bearing: a later entry may bind onto a table an earlier one made. **A refused entry stops the run and the entries before it stay** — they are separate copies, committed as they land — so the refusal carries the server's own message plus what already landed and the one-package command to retry with. `--json` prints one object and nothing else (`entities`, each with `bound_by`: `created` by this run, `id` through the binding, or `label` for an alias nothing had bound; `roles`, `record_ids`, `rows_skipped`, `drift`, `applied: [{package, apps}]` — always present, empty included, so a reader cannot mistake "applied nothing" for "too old to say" — plus `organization_id`/`workspace_id` and a `warnings` array). Admin-only. A model STATES its apps and `apply` builds none of them — `lotics app create --from <model.json>#<app>` or `lotics workspace build` does; a published package named in `apply` brings its own. |
 | `lotics scaffold apply <model.json> --documents` | **The `files` cells of `rows`, and nothing else** — no table, no field, no row. The half a re-apply cannot redo: rows land only into empty tables, so a model whose binaries were missed the first time has no other way back, and a model whose records are pictured ships with empty pictures until this runs. Each distinct path is uploaded once, recorded against the file it became, and attached with `update_records add_to` onto the record the binding names — joined by the row's own REF (`<entity>:<ref>`), so a row added to or removed from the file since the apply changes nothing about the rest. **A ref this workspace holds no record for is a refusal**, naming it: the alternative is filing its document under whichever row happened to sit at that position. **Idempotent by construction**: an uploaded path is reused from the binding's `documents` map and `add_to` is a set union, so running it twice leaves one id in the cell. The field is bound by the id the binding recorded, and by LABEL only for a field added since — a label that has moved with nothing bound to it is refused, pointing at `lotics scaffold diff`. Admin-only. |
 | `lotics scaffold diff <model.json>` | **Where the file and this workspace have come apart.** Runs `scaffold export` against the selected workspace and prints what the model has and the workspace lacks, the reverse, and every field the two disagree about — a type, the option labels one carries and the other does not, where a new row starts (`default`, none where the model states none) and how its values read (`format`, the type's plain one where none is stated; a date's is named and left, since `apply` would rewrite its dates; a number's `unit`, one moved between two of one dimension named and left, since `apply` would convert its figures), or a PAIRING the model declares over a link that is one-way here, which `apply` refuses and nothing else named — and each `unique` set one side declares and the other lacks, compared as an unordered set of the workspace's fields. **Joined on the ids this workspace remembers**, entity then field then select option, and on LABEL only for an alias nothing has bound — a new entity, or a workspace this model has never been applied to, which is the join the first apply itself makes. A relabel on either side therefore prints as one rename under the alias (`order.state: label: model "Stage" · workspace "Giai đoạn"`) rather than as one thing the workspace lacks and one the model lacks, which is what made an additive apply add a second one. A bound id this workspace no longer serves is named with the command that puts it back, never silently re-bound by label. **Exits 1 on any difference**, so it is a gate: a starter published from a model that has drifted would ship the FILE's labels while the workspace uses others, and nothing else compares them. Checks the file offline first. Admin-only (the export is). |
@@ -68,9 +68,9 @@ Per-command syntax, flags, contracts, and gotchas for the public `lotics` CLI. S
 | `lotics docs` \| `lotics docs <area>[/<section>]` | The index of the reference docs, **resolved out of the packages installed beside this project** — the model reference alone is carried by this CLI (`lotics docs model`). **Both levels are discovered by looking**: every `@lotics/*` package carrying an `AGENTS.md` or a `docs/` in any `node_modules/@lotics` from the current directory UPWARD (nearest wins, so a hoisted root copy never shadows the one a project's own imports resolve to), and within each, every area it actually ships. Titles come from each file's own `# heading` and the version from the installed `package.json`, so a doc OR a whole package added upstream appears with no change to this CLI, and a skewed install is visible rather than reassuring. A package's index is named after the package (`lotics docs ui`), never `index`. `@lotics/app-runtime`, `@lotics/ui` and `@lotics/cli` sort first as a reading ORDER, not a filter. **Output is ONE PAGE, 16 KB, navigation included**: a doc that does not fit prints its opening and the addresses that reach into it — its sections with their sizes, or, for a reference that is one table (this file, the kit's catalog), the name of every row. `lotics docs <area>/<section>` prints that section, a path of sections (`model/field/select`) finds each inside the one before, and `lotics docs ui/catalog/Button` that one row; a unique prefix is enough, and an address matching two parts is refused with both. Both levels print to **stdout** — the index is the payload of a bare `lotics docs`, so `lotics docs | grep -i excel` works — with only the provenance line on stderr; a name two packages share is refused with both qualified forms (`lotics docs ui/templates`) rather than resolved silently. Needs no auth. Outside a project only `@lotics/cli`'s own resolve, and it says so. |
 | `lotics docs model` \| `lotics docs model/<section>[/…]` | **The model reference, from inside the binary** — the one doc this CLI carries rather than resolves, because it describes this CLI's own model checker; listed first by `lotics docs`, at this CLI's version. Its first page is what a model composes with, the working order (jobs → entities and fields → `records` → one app per job → `scaffold check` → `app preview` → `workspace build`) and the section addresses; every page of it is whole. Every top-level key of a `model.json`, every field `type` the contract admits with the config each one needs, the option / view / role / inline-template shapes, `records` (how a row of each entity is recognised), `write_rules`, `apps` (each register, record, act and check), the row format (relative dates `@today` / `@month-start` with whole-day offsets; links as `"<entity-alias>:<ref>"`), the rules, the `apply` list (packages copied in after the model's own tables, each with an optional `bind` onto them), the `preset` block (a published model's branches and its at-most-two questions), the **`from` form** — `{from, variants, rename, entities, rows, records, write_rules, apps, apply}`, which names a preset by SLUG instead of restating it — and one complete worked example. **Offline, no account.** |
 | `lotics report '<json>'` \| `lotics report @report.json` | File a report with the Lotics team about what got in your way. **Covers the classes telemetry structurally cannot see**: a capability that does not exist (no command ran, so nothing was recorded), a command that exited 0 having done the wrong thing, an error whose message did not name the remedy, and anything that made authoring slower than it should be. **A frame, not a paragraph** — `{goal, actual, expected?, tried?, wanted?}`, `goal` and `actual` required, unknown keys dropped rather than refused. **No severity or category.** Ingest is inline JSON, `@file`, or `-` for stdin. A bare sentence is refused with the frame printed beside it, so the fix is one step; a bare invocation prints the frame BEFORE asking for a credential, since someone whose key will not resolve is exactly who has something to report. **Not spooled**: unlike telemetry it posts inline, prints whether it landed, and exits non-zero if it did not, echoing the report back so a failed send never loses it. Runs regardless of `LOTICS_TELEMETRY` — invoking it IS the consent that passive collection needs an opt-in for — but with telemetry off there are no recorded commands to attach, and it says so rather than implying context it does not have. Requires auth. Never paste records, file contents, or credentials. **Prints the id of each frame filed** — a filing nobody can cite cannot be answered about. The ids come from the server, so an instance that only logs the frames prints the count alone; the CLI never mints one of its own, which would hand back a token that resolves to nothing. |
-| `lotics app check` | Every pre-flight `deploy` runs, WITHOUT building or shipping. **First, whether this project is even based on the served version** — the one thing a deploy REFUSES outright rather than pushing (the server 409s a stale `prev_version_id`), and the one finding that invalidates every other: a stale tree and the live app are two different apps, so comparing them reports nothing trustworthy. Stale exits 1 naming both versions and stops before the rest; a project with no stamp at all — or an app with no version yet — is a first deploy, not a conflict. `deploy` runs the SAME assertion, so a stale tree fails before it pushes or builds. Then: the manifest's agent schemas against the live app row, every binding a deploy would push, aliases the source calls that nothing bound (queries, workflows AND agents), bindings this bundle stopped calling (the same transition — and the same baseline — `deploy` reports, so the two cannot disagree), capability-gated SDK calls the manifest doesn't declare, a missing icon/theme, a missing app `description` (it heads the capability catalog the chat agent reads every turn, and its absence has no other symptom), a `vite.config.ts` that never defines `global`/`__DEV__` **in a project that still ships react-native** (read off its own `package.json` — an app on the React DOM kit bundles none of it and the check would be advice to define two globals nothing reads), and a `window.open` in the app's own source (each fails ONLY in the deployed app: dev bundles with esbuild and production with rollup, so typecheck, lint, build and `app dev` are all green), an agent whose capability and its reach disagree, in EITHER direction, over any of the five declaration-bound tools (`run_app_query`/`run_app_workflow` against `query_aliases`/`workflow_aliases`; `grep_knowledge`/`read_knowledge`/`list_knowledge` against `knowledge_doc_ids`) — the tool is the capability, the list is the reach, and a tool with no reach means every call it makes is refused while the run still COMPLETES, so it surfaces as a model ignoring its prompt; read off the live row, never the manifest, which mirrors those fields but is pushed by no verb, and a notice for any alias the source computes at runtime (invisible to every check here and to `--prune`'s unbind guard). **And whether the runtime this app builds against has fallen behind what is published** — `@lotics/app-runtime` alone, since the kit an app lists sits at the runtime's range, read from `node_modules` rather than the range because a caret is minor-locked below 1.0 (`^0.13.x` can never resolve `0.14`, and `npm update` does nothing). A release LINE behind is loud and names `lotics app kit --published` and `@lotics/ui`'s `MIGRATION.md`; anything smaller is one quiet `npm update` line, because a warning that fires on every deploy is one the reader stops seeing. An app still listing `@lotics/app-sdk` is told it moved into the runtime and which verb migrates it. The lookup is bounded and every failure is silence: a version check must never become a new way for a deploy to fail. Every deploy finding is the same helper `deploy` calls, so a green check means a deploy will not complain. **And the PORTABILITY gate, one of three rules `check` runs that a deploy does not** (the others are the undeclared call site below and `package.json#lotics.writes`/`deletes`, held against what every workflow body writes and deletes) — the ids an app cannot carry into another workspace, over the working tree, with the same exclusions the deploy tar applies. Two rules. **An id this workspace MINTED**, written into `src/`, a `.md` or the app's own docs — it resolves to nothing in a copy, and in prose it is an instruction the copier's agent follows; this is the one a `library publish` also refuses, on the uploaded archive. **And an id-shaped STAND-IN** too short for the generator that mints its prefix (`"opt_X"`, `"fld_a"` — quoted or in a code span, so a bare `opt_in` stays legal, and never in a test file), which only `check` runs, and which additionally reads a workflow body and the manifest: those two are exempt from the first rule because a publish INVERTS a real id there and cannot invert a fake. **And an `app_` id anywhere in `app.json`**: the spec names no other app, so a pasted one is re-pointed by nothing. Each is reported as `<file>:<line> — <id>` with the one edit that fixes it. This gate reads only the files, but the command around it still needs a resolvable credential and the live app row, so it is not an offline check. **And every bound workflow body, checked as `app workflow check` checks it** — the same isolated per-alias program against the pulled `.lotics/workflows/<alias>.globals.d.ts`, then the server's `verify_only` verdict on each body that passed — and on each body no local pass could type (no globals, no `typescript`), since that is what a push would meet — which exits 1 on any issue and warns when the server is not reached. The server verifies a body once, at the save that wrote it, so a body whose declared types have since moved stays stored, matching what is live, and is refused by the next writer — a starter copy, in somebody else's workspace. The verdict is as fresh as those types, which `pull`, `workflow pull` and `codegen` refresh. **And the app's own `npm run typecheck`**, after regenerating the `.lotics/*.d.ts` companions from the manifest — the same run a deploy makes before building, so a filter or sort key the query does not project fails here rather than at the first member's request. **Exits 1 on that, on a body the types refuse, on a failing typecheck, and on what a `deploy` would REFUSE** — **every manifest query the server would refuse** — held to the server's own query gate (the declaration, each query it runs as, the tables it reads through `get_table`, its filters and the columns it outputs; the owner's reach and the SQL compile stay the server's), in the server's words — and a query or workflow `description` over the 300-character capability cap, which is a binding no `set` and no deploy will take (the rule is `@lotics/shared`'s, the same one the server refuses with, and `workflow set` / `query set` ask it before sending anything — so a long line costs one edit rather than a failed push per alias). **And every manifest query declared with NO description**, named in one line: that sentence is what a chat or MCP caller chooses between aliases by, and an alias is a JS identifier. `app create --from` deliberately writes none — a template over a shape's own English reads like a line about the business while saying nothing — so a generated app is told once, here, which lines are the author's to write. Both are things a deploy would refuse, so CI gating on a green check means a deploy will not refuse. **Every binding the project has ahead of the app** — an agent schema that disagrees with the live app, an edited workflow body or declaration, edited agent prose, a changed query — **is named as what the deploy will push**, each with the one-alias `set` verb, and never fails the check: the deploy pushes it itself, so a check red over it disagreed with the deploy it stands in for (and with the check `app regenerate` runs). Genuine advisories (capabilities, branding, a runtime-computed alias, orphaned bindings) stay advisory and never fail it. **`--screens` adds the rendered surface, and is its own entry below.** It runs after the typecheck and only when it passed — a type error renders nothing to measure — and its findings, and any write it refused, fold into this command's exit code. **It also refuses a call site the manifest no longer declares.** `useQuery`/`useWorkflow` keep a bare-string overload for a computed alias, so deleting or renaming an alias leaves every call site compiling and failing only when the screen renders — the one edit most likely to orphan a call site is the one the generated types cannot catch. The alias literals in `src/` are matched against `package.json#lotics.queries`/`.workflows` (the manifest, not the live row: the server still SERVES an alias whose declaration was just deleted, because a deploy never unbinds), and an undeclared one exits 1. `app codegen` prints the same finding as a warning, since it is the command an author runs right after editing the manifest. **A failing body that is byte-identical to the one the server is running is labelled as such**: its `lotics.synced.workflows.<alias>.content` baseline proves the file has not been edited since it was pushed or pulled, so the failure is a grammar migration the stored body is owed rather than a stale checkout — a link field reads as an id array, so drop `.id` or descend with `linked(…)`. Until the body is edited and `set`, the stored one keeps running as it always has. **And a query that reads past a table's ROW RULE.** A table's `private_filters` bind the CALLER, and an app query's caller is the app's OWNER — the viewer needs no table access, `app:use` is the grant — so the rule passes and every row is served. For each table a declared query reads (`GET /v1/tables/{id}`, the one surface that serves the rule), the viewer predicate — `current_member in_any_group`, or a member field's `is_current_member` / `is_not_current_member` — is attributed to the SCAN it guards: a `from_table`'s own `filter`, or an enclosing `filter` node's predicate, whose rows are the ones that scan produced. So a clause written over one table never silences the finding for the ruled table joined beside it, and every scan no clause covers is named with its table. A table this credential may not read (403, or 404 for one that is gone) is dropped — a rule that cannot be read is not evidence of one — while any other failure of that read fails the command, since "I could not ask" must never render as "there is no rule". Advisory, never part of the exit code: an app that deliberately serves the whole table to a desk of people who may all read it is legitimate, and nothing here can tell the two apart. **And it names the workflows a chat or MCP caller is offered with no description** (one `get_app_capabilities` read, the reader's own view of the app): that text is what those callers choose between aliases by, and without it they choose by the alias. It cannot be counted from the manifest — a workflow bound out of band is not declared there at all. Advisory, never part of the exit code. **It regenerates `.lotics/app_fields.ts` from the live schema before typechecking**, as a deploy does — a gitignored map from a moved schema otherwise passes. **And three claims the app makes**: a bound body's writes against `package.json#lotics.writes`, read as the step tree the server would store, exit 1 on a field no entry covers, and declaring none only warns; a description carrying `<placeholder>` syntax exits 1, since the catalogue escapes angle brackets — state the format as an example; so does one naming a desk `query_apps` does not list. |
-| `lotics app check --screens [--screen <label>] [--width <n>] [--shots <dir>] [--changed]` | **`--screens` adds the rendered surface**: the app is served the way `app dev` serves it (its real data, this key), rendered headless in Chrome (`CHROME_PATH`/`LOTICS_CHROME`, then Playwright's, then system) at 1280 and 375. **The screens are its navigation's destinations** — a `nav` landmark's `a[href]` or `role="link"` (an app's route lives in its router, so the kit's shell renders each screen as a button carrying the link role and no `href`), else the first tab strip, else the root — in that order, because a screen's own lifecycle desk draws a tablist too, and reaching for one first walks a screen's STAGES as if they were the app's. **Each screen is then WALKED THROUGH ITS OWN DOORS**, nothing configured per app: every door is named by something the document states. A record: a row stamped `data-opens="page"`, a real `a[href]`, or an EMPTY DOOR — a named control with no text and no child element, the only legal whole-row press target, which reaches a drawer register, a Schedule row and a hand-written row alike. On a record — a page, or a drawer standing over its register, whose doors are the drawer's and never the register's behind it: the acts menu (`aria-haspopup`), the first child row, and every tab (a drawer's sections, the companion) and every section that is a page of its own, each measured and walked for the dialogs its acts and adds raise, a dialog the record already opened not twice. On any surface: every dialog a visible primary or secondary act raises — nothing announces one, so a press is kept only where an overlay appeared. **THE RUN IS A READ, AND THE NETWORK IS WHERE THAT IS ENFORCED — never a rule about what the page may draw.** The app frame holds no credential and Chrome runs on a throwaway profile, so every call the app makes reaches the workspace through this CLI and nowhere else, and this CLI serves an ALLOWLIST of reads. Everything outside it is refused before the call leaves the machine — a workflow run, a record write, an upload, an agent run, a comment, and any op this CLI does not know, which is refused because it is not on the list rather than because anyone listed it. A refused call HEADS the report, naming the surface, the control the app had focused and the RPC by its alias (never a payload value), and fails the run on its own: a walk that provoked a write left the app in a state no reader could have put it in, so nothing measured under it means anything. What the walk does is bounded on the page as well: nothing inside an open overlay but the record drawer it opened, nothing typed, no submit and nothing in a form, no act the kit marks costly (`data-tone` danger/warning) and none whose own name is the write, in either language — and nothing is FOCUSED, because focus cannot be taken from one field without leaving another, and a field that saves itself when the reader leaves it saves on exactly that. Where a reading needs the focused state, as the focus-ring rule does, Chrome is asked to PAINT `:focus-visible` and release it again: the cascade answers, focus never moves and no event is dispatched. Each overlay closes with Escape, confirmed closed; a record is descended at most twice; the walk stops at twenty-four surfaces per screen and NAMES each door it left. Each surface is measured once no request is in flight, and the measurable probes of `@lotics/ui` docs/reviewing.md run over the DOM, each finding printing the rule it IS — its law, its section and the one edit that answers it — so the numbers need no key. What they exempt is what the screen itself declares: a register's ordinal gutter (a column counting to the row count is the shape's numbering, which no app can treat), a strip whose list carries `data-order="sequence"` (a lifecycle rail, which composition.md permits under a screen's tabs), a hairline or `clip-path`-clipped leaf (the visually-hidden node a control plants for a screen reader), and a leaf whose own computed line clamp states a count over a sentence. **A meter counts as an encoding only where it draws a POSITION** — `aria-valuenow` inside a range with room left. A bar pinned at its own maximum — what a meter alarmed AT its maximum draws on every alarmed row — reads the same as every value above it, and one with no maximum states none; both are counted in the census's `devices` and out of its `encoded`, so "nothing drawn" and "drawn and saying nothing" never read alike. The bare values that remain are grouped into columns, each named by the heading over it, so a finding says WHICH slot draws its figures as words. **Two finding classes read what geometry cannot.** `clutter` (docs/hierarchy.md): a second primary act, a value in two places on a record, a box reserving more lines than it holds, markers on over half a form's fields, a second accent. `right_form` (docs/screen.md's device index): a boolean as a two-option select, a day run as a repeated date column where the kit ships `Schedule`. Each prints a line per rule fired, with three offenders. **Four read room and counts** (reviewing.md §8l–§8o): `reserved_blank`, a register column 40px or wider that is blank (no text, no control) on more than half the rows in view, or a row-act gutter leaving 40px beside the acts on them — a row a complete register draws for nothing stored (`data-ghost`, a day nobody filed) is no row of the set; `shed_with_room`, a column the register's width gave up (the kit states them as `data-shed`) that, with the gap between two columns, fits in what its slots hold beyond what each must seat — its heading and widest ink, a fixed column's whole width, the acts most rows have; `action_off_heading`, a section's create (a control stating `data-create`) drawn in its body rather than its heading row; and `partial_count`, a count or `n of N` equal to the rows in hand of a read the server cut on the screen being walked, where the server's own count of the same alias, params and filter — asked by this CLI, a read — says otherwise. **Three read a record's anatomy** (docs/hierarchy.md), over each record on the surface with the title heading it: `record_primary`, a primary act in a block's heading row — a block's add is secondary beside the record's own acts (two under one heading are `clutter`'s); `fact_repeats`, a field whose value reads the same as the header's own words, whatever its case; and `empty_block`, a block that draws nothing under its heading, not even the line saying it is empty. A census line per screen (text runs, money strings, bare values against the devices reading, tab strips) prints first, so a clean verdict over a screen that rendered nothing cannot pass; a screen that renders no text is itself a finding, and one still changing after fifteen seconds is measured as it is. **A cold dependency optimisation is waited out, not measured**: the first paint has its own bound, far longer than settle's, since an app's modules load after its document completes and Vite holds them; only a MOUNTED, idle, textless frame is blank at once, and the finding names the wait and its blocker. It also RELOADS the page under the probe, so a width is measured again once; a second reload is a page that keeps moving and fails. **`--screen <label>` and `--width <n>` narrow a run** (repeatable, comma-separated; the label matches case-insensitively as a substring), for the author iterating on one screen who would otherwise pay a typecheck, a Vite boot and every screen at both widths on every edit; the clean verdict then names only the widths covered, and a `--screen` matching nothing is refused. **`--shots <dir>` writes what the run measured** — a `<surface>@<width>.png` and a `<surface>@<width>.json` per surface, off the SAME settled frame the probes read, so a shot and a finding can never describe different pixels. The PNG is the WHOLE surface: an app scrolls inside a box of its own, so the window is grown to the height the probe measured and put back, and an overlay a resize dismissed is shot as the viewport, the sidecar saying so (`app dev`'s header band is in it — the band the app was laid out under). The JSON is the half a picture cannot carry: the nav's first item's left edge, the title and first section heading, the primary acts by name, label/value pairs against what the folds state, values drawn twice, money that wraps, text the layout cut, and the console errors and uncaught exceptions the frame raised. Named `<nn>-<slug>` plus one `__<step>` per door taken (`__record`, `__menu`, `__child`, `__fold`, `__dialog-<n>`, `__tab-<its name>`, `__section-<its name>`); `<nn>` is the walk's index, which lists the directory in reading order and keeps two labels that fold to one ASCII slug apart. The row a record surface opened from is the sidecar's `opened_from` and that screen's census line, never a file name — it is a person's data. The directory is created if missing, and refused before the dev server boots when it cannot be; `--shots` without `--screens` is refused. Findings exit 1 like the rest. **Where it renders**: the app's files are copied into the render project of its DEPENDENCY SET — `~/.lotics/render/<key>/`, keyed by the manifest's `dependencies` and `devDependencies` as written, a `file:` tarball by its bytes — the one `app preview` renders in, so the install and Vite's dependency optimisation are paid once per set rather than per app; one render holds a project at a time (`<key>.lock` — a second run waits, naming the app and pid, and takes over a dead hold). Where that install resolved another runtime or kit than the app's own, the run says which, since a deploy builds the app's own. The run ends with what each phase cost: `preflight · install|reuse · walk`. **`--changed`** walks only when something the screens READ has moved since this checkout's last walk — each part of `app.json` (every record apart), each declared query, each bundled source file, the field map, the ranges, the kit the render resolved and the CLI version, whose probes measure them — read against `.lotics/screens_check.json`, which every walk writes; with nothing moved, that walk's report is printed again under its timestamp and still fails the run. An app's screens all read its one spec, so a move walks the app whole. It walks, saying why, with no earlier walk, a different `--screen`/`--width`, or `LOTICS_UI_SRC` set (a linked working copy has no fingerprint); the workspace's rows are not an input. `--changed` without `--screens`, or beside `--shots`, is refused. |
-| `lotics app preview <model.json>#<app> [--shots <dir>] [--width <n>] [--screen <label>] [--kit <path>]` | **The app a model states, RENDERED — before a table exists, and with no credential in the process.** The model is checked offline, the named app is compiled against the workspace the model WOULD become (synthetic `tbl_`/`fld_`/`opt_`/`grp_` ids, minted positionally off the file), and every read the app makes is answered from the model's own `rows` — projected under the columns the query names, narrowed by the params it declares and sorted the way it states, so a child block under a record holds that record's rows and not every parent's. An entity the app reads that states no `rows` gets three synthesized, so a register never measures clean over nothing. Then the SAME headless walk `app check --screens` takes: the register, the record its first row opens in its door, and the add dialog, at 1280 and 375, measured by the same probes. `--shots <dir>` writes a PNG and a sidecar per surface into `<dir>/<app>/`, so apps previewed into one directory never overwrite each other. **Exits 1 on any finding**, exactly as the check does, and prints what each phase cost — bind, install-or-reuse, walk. **Nothing is created and nothing is reached** — no app row, no table, no workspace; the only network it needs is the npm registry, and only when the kit has moved. It renders in the cached render project of its dependency set (`~/.lotics/render/<key>/`, shared with `app check --screens` and with every model whose app lists the same set, and held by one render at a time — a second run waits, naming the app and pid holding it, and takes over a hold whose process is gone): the runtime and its kit are PUBLISHED packages, so a project is what a preview needs to install them into — the CLI has no renderer of its own. **It installs the runtime this CLI was built for**, and the kit through it, never `latest`: the spec this binary's generator writes is the one that runtime reads, and a version that published mid-session is a runtime nobody checked the model against. A version the registry does not serve yet is refused before npm runs, naming `--kit` — needed only until that runtime is published. **`--kit <checkout>`** (repeatable) — a checkout's ROOT, which names the app-runtime and the ui it holds, or one of those two packages — renders against that checkout instead — built, packed and proven the way `lotics app kit` installs one into an app — and the run prints that it rendered against a LOCAL CHECKOUT, first and last, so the render is never read as the published one. The project is a CACHE and never a source: `npm install` runs on the first render of a set and again only when a checkout's tarball did, and every file in it is rewritten from the model on each run. **A document a row names by path is served** from the project's own static root, as the named file a workspace would hold, so a files block and a record's picture read what the model states; a `fil_` id is carried by name. **`--record <ref>`** opens that row's record by its own address — a row of the app's entity, by the ref the model gives it, open or closed — where the walk otherwise opens the first row the register lists; a ref naming no such row is refused with the refs there are. **Every read is answered at the BRIDGE**, not inside the page. The app SDK's design-time fixture (`registerMockFixture` + `?__mock=1`) is the wrong half of this: the flag lives in the app's own url, and the first record page is a navigation the app's router performs — so from that surface onwards the flag is gone and every read falls through anyway. The bridge is the one place every call arrives whatever the url says, so `query`, `field_options`, `members` and `context` are answered there from one reading of the rows, and the generated entry ships EXACTLY as `app create --from` writes it. **A preview writes nothing**: a workflow, an upload or an agent run meets the same read gate `app check --screens` uses, so the app draws its own refusal path rather than reporting a save nothing moved for, and the call is reported above the census the way the check reports one. **What came out empty, it names**: a `formula`, `rollup`, `lookup` or `autonumber` is computed here rather than stated in the model, and the census lists each computed column that stayed empty on EVERY row of a table that has rows — named by outcome, since a total drawn blank is either the app's own answer or a hole and only the run can say which. The census reads the CELLS and not the field types, so nothing is exempt by kind. **A computed column that came out `#ERROR:` is named too, and the run refuses it** before anything is installed: a wall of red measures clean, and a run that drew it and exited 0 would tell its author the app is fine. **It computes in UTC**, because a model states no timezone — a row's `@today`, an autonumber's `{YEAR}` and a record's own clock all land on the run's own date at UTC midnight, so two runs of one model draw the same register wherever they are made. |
+| `lotics app check` | Every pre-flight `deploy` runs, WITHOUT building or shipping. **First, whether this project is even based on the served version** — the one thing a deploy REFUSES outright rather than pushing (the server 409s a stale `prev_version_id`), and the one finding that invalidates every other: a stale tree and the live app are two different apps, so comparing them reports nothing trustworthy. Stale exits 1 naming both versions and stops before the rest; a project with no stamp at all — or an app with no version yet — is a first deploy, not a conflict. `deploy` runs the SAME assertion, so a stale tree fails before it pushes or builds. Then: the manifest's agent schemas against the live app row, every binding a deploy would push, aliases the source calls that nothing bound (queries, workflows AND agents), bindings this bundle stopped calling (the same transition — and the same baseline — `deploy` reports, so the two cannot disagree), capability-gated SDK calls the manifest doesn't declare, a missing icon/theme, a missing app `description` (it heads the capability catalog the chat agent reads every turn, and its absence has no other symptom), a `vite.config.ts` that never defines `global`/`__DEV__` **in a project that still ships react-native** (read off its own `package.json` — an app on the React DOM kit bundles none of it and the check would be advice to define two globals nothing reads), and a `window.open` in the app's own source (each fails ONLY in the deployed app: dev bundles with esbuild and production with rollup, so typecheck, lint, build and `app dev` are all green), an agent whose capability and its reach disagree, in EITHER direction, over any of the five declaration-bound tools (`run_app_query`/`run_app_workflow` against `query_aliases`/`workflow_aliases`; `grep_knowledge`/`read_knowledge`/`list_knowledge` against `knowledge_doc_ids`) — the tool is the capability, the list is the reach, and a tool with no reach means every call it makes is refused while the run still COMPLETES, so it surfaces as a model ignoring its prompt; read off the live row, never the manifest, which mirrors those fields but is pushed by no verb, and a notice for any alias the source computes at runtime (invisible to every check here and to `--prune`'s unbind guard). **And whether the runtime this app builds against has fallen behind what is published** — `@lotics/app-runtime` alone, since the kit an app lists sits at the runtime's range, read from `node_modules` rather than the range because a caret is minor-locked below 1.0 (`^0.13.x` can never resolve `0.14`, and `npm update` does nothing). A release LINE behind is loud and names `lotics app kit --published` and `@lotics/ui`'s `MIGRATION.md`; anything smaller is one quiet `npm update` line, because a warning that fires on every deploy is one the reader stops seeing. An app still listing `@lotics/app-sdk` is told it moved into the runtime and which verb migrates it. The lookup is bounded and every failure is silence: a version check must never become a new way for a deploy to fail. Every deploy finding is the same helper `deploy` calls, so a green check means a deploy will not complain. **And the PORTABILITY gate, one of three rules `check` runs that a deploy does not** (the others are the undeclared call site below and `package.json#lotics.writes`/`deletes`, held against what every workflow body writes and deletes) — the ids an app cannot carry into another workspace, over the working tree, with the same exclusions the deploy tar applies. Two rules. **An id this workspace MINTED**, written into `src/`, a `.md` or the app's own docs — it resolves to nothing in a copy, and in prose it is an instruction the copier's agent follows; this is the one a `library publish` also refuses, on the uploaded archive. **And an id-shaped STAND-IN** too short for the generator that mints its prefix (`"opt_X"`, `"fld_a"` — quoted or in a code span, so a bare `opt_in` stays legal, and never in a test file), which only `check` runs, and which additionally reads a workflow body and the manifest: those two are exempt from the first rule because a publish INVERTS a real id there and cannot invert a fake. **And an `app_` id anywhere in `app.json`**: the spec names no other app, so a pasted one is re-pointed by nothing. Each is reported as `<file>:<line> — <id>` with the one edit that fixes it. This gate reads only the files, but the command around it still needs a resolvable credential and the live app row, so it is not an offline check. **And every bound workflow body, checked as `app workflow check` checks it** — the same isolated per-alias program against the pulled `.lotics/workflows/<alias>.globals.d.ts`, then the server's `verify_only` verdict on each body that passed — and on each body no local pass could type (no globals, no `typescript`), since that is what a push would meet — which exits 1 on any issue and warns when the server is not reached. The server verifies a body once, at the save that wrote it, so a body whose declared types have since moved stays stored, matching what is live, and is refused by the next writer — a starter copy, in somebody else's workspace. The verdict is as fresh as those types, which `pull`, `workflow pull` and `codegen` refresh. **And the app's own `npm run typecheck`**, after regenerating the `.lotics/*.d.ts` companions from the manifest — the same run a deploy makes before building, so a filter or sort key the query does not project fails here rather than at the first member's request. **Exits 1 on that, on a body the types refuse, on a failing typecheck, and on what a `deploy` would REFUSE** — **every manifest query the server would refuse** — held to the server's own query gate (the declaration, each query it runs as, the tables it reads through `get_table`, its filters and the columns it outputs; the owner's reach and the SQL compile stay the server's), in the server's words — and a query or workflow `description` over the 300-character capability cap, which is a binding no `set` and no deploy will take (the rule is `@lotics/shared`'s, the same one the server refuses with, and `workflow set` / `query set` ask it before sending anything — so a long line costs one edit rather than a failed push per alias). **And every manifest query declared with NO description**, named in one line: that sentence is what a chat or MCP caller chooses between aliases by, and an alias is a JS identifier. `app create --from` deliberately writes none — a template over a shape's own English reads like a line about the business while saying nothing — so a generated app is told once, here, which lines are the author's to write. Both are things a deploy would refuse, so CI gating on a green check means a deploy will not refuse. **Every binding the project has ahead of the app** — an agent schema that disagrees with the live app, an edited workflow body or declaration, edited agent prose, a changed query — **is named as what the deploy will push**, each with the one-alias `set` verb, and never fails the check: the deploy pushes it itself, so a check red over it disagreed with the deploy it stands in for (and with the check `app regenerate` runs). Genuine advisories (capabilities, branding, a runtime-computed alias, orphaned bindings) stay advisory and never fail it. **`--screens` adds the rendered surface, and is its own entry below.** It runs after the typecheck and only when it passed — a type error renders nothing to measure — and its findings, and any write it refused, fold into this command's exit code. **It also refuses a call site the manifest no longer declares.** `useQuery`/`useWorkflow` keep a bare-string overload for a computed alias, so deleting or renaming an alias leaves every call site compiling and failing only when the screen renders — the one edit most likely to orphan a call site is the one the generated types cannot catch. The alias literals in `src/` are matched against `package.json#lotics.queries`/`.workflows` (the manifest, not the live row: the server still SERVES an alias whose declaration was just deleted, because a deploy never unbinds), and an undeclared one exits 1. `app codegen` prints the same finding as a warning, since it is the command an author runs right after editing the manifest. **A failing body that is byte-identical to the one the server is running is labelled as such**: its `lotics.synced.workflows.<alias>.content` baseline proves the file has not been edited since it was pushed or pulled, so the failure is a grammar migration the stored body is owed rather than a stale checkout — a link field reads as an id array, so drop `.id` or descend with `linked(…)`. Until the body is edited and `set`, the stored one keeps running as it always has. **And a query that reads past a table's ROW RULE.** A table's `private_filters` bind the CALLER, and an app query's caller is the app's OWNER — the viewer needs no table access, `app:use` is the grant — so the rule passes and every row is served. For each table a declared query reads (`GET /v1/tables/{id}`, the one surface that serves the rule), the viewer predicate — `current_member in_any_group`, or a member field's `is_current_member` / `is_not_current_member` — is attributed to the SCAN it guards: a `from_table`'s own `filter`, or an enclosing `filter` node's predicate, whose rows are the ones that scan produced. So a clause written over one table never silences the finding for the ruled table joined beside it, and every scan no clause covers is named with its table. A table this credential may not read (403, or 404 for one that is gone) is dropped — a rule that cannot be read is not evidence of one — while any other failure of that read fails the command, since "I could not ask" must never render as "there is no rule". Advisory, never part of the exit code: an app that deliberately serves the whole table to a desk of people who may all read it is legitimate, and nothing here can tell the two apart. **And a scan no index can narrow**, on a table of 2,000 records or more: each `from_table` is judged with only the required params set, then with each optional param its filter reads set alone, pruned as the runtime prunes an omitted param, and named with the params that leave it unserved and the conditions at fault. A filter is served when it cannot hold without an index-served condition — any branch of an AND, every branch of an OR — or its source carries a `search`; an index serves an exact match (`equals`/`is_any_of` on text, comparisons on numbers and dates, built at deploy) `has_any_of`/`has_all_of` on a select, member or link, and `is_current_member`, never a substring match. A call whose filter keeps no condition asked for every record and is not named; a scan beneath a `filter` over its projection is not judged, since the compiler may move that filter into it. A push prints the same advisory for the queries it sends. Advisory, never part of the exit code. **And it names the workflows a chat or MCP caller is offered with no description** (one `get_app_capabilities` read, the reader's own view of the app): that text is what those callers choose between aliases by, and without it they choose by the alias. It cannot be counted from the manifest — a workflow bound out of band is not declared there at all. Advisory, never part of the exit code. **It regenerates `.lotics/app_fields.ts` from the live schema before typechecking**, as a deploy does — a gitignored map from a moved schema otherwise passes. **And three claims the app makes**: a bound body's writes against `package.json#lotics.writes`, read as the step tree the server would store, exit 1 on a field no entry covers, and declaring none only warns; a description carrying `<placeholder>` syntax exits 1, since the catalogue escapes angle brackets — state the format as an example; so does one naming a desk `query_apps` does not list. |
+| `lotics app check --screens [--screen <label>] [--width <n>] [--shots <dir>] [--changed]` | **`--screens` adds the rendered surface**: the app is served the way `app dev` serves it (its real data, this key), rendered headless in Chrome (`CHROME_PATH`/`LOTICS_CHROME`, then Playwright's, then system) at 1280 and 375. **The screens are its navigation's destinations** — a `nav` landmark's `a[href]` or `role="link"` (an app's route lives in its router, so the kit's shell renders each screen as a button carrying the link role and no `href`), else the first tab strip, else the root — in that order, because a screen's own lifecycle desk draws a tablist too, and reaching for one first walks a screen's STAGES as if they were the app's. **Each screen is then WALKED THROUGH ITS OWN DOORS**, nothing configured per app: every door is named by something the document states. A record: a row stamped `data-opens="page"`, a real `a[href]`, or an EMPTY DOOR — a named control with no text and no child element, the only legal whole-row press target, which reaches a drawer register, a Schedule row and a hand-written row alike. On a record — a page, or a drawer standing over its register, whose doors are the drawer's and never the register's behind it: the acts menu (`aria-haspopup`), the first child row, and every tab (a drawer's sections, the companion) and every section that is a page of its own, each measured and walked for the dialogs its acts and adds raise, a dialog the record already opened not twice. On any surface: every dialog a visible primary or secondary act raises — nothing announces one, so a press is kept only where an overlay appeared. **THE RUN IS A READ, AND THE NETWORK IS WHERE THAT IS ENFORCED — never a rule about what the page may draw.** The app frame holds no credential and Chrome runs on a throwaway profile, so every call the app makes reaches the workspace through this CLI and nowhere else, and this CLI serves an ALLOWLIST of reads. Everything outside it is refused before the call leaves the machine — a workflow run, a record write, an upload, an agent run, a comment, and any op this CLI does not know, which is refused because it is not on the list rather than because anyone listed it. A refused call HEADS the report, naming the surface, the control the app had focused and the RPC by its alias (never a payload value), and fails the run on its own: a walk that provoked a write left the app in a state no reader could have put it in, so nothing measured under it means anything. What the walk does is bounded on the page as well: nothing inside an open overlay but the record drawer it opened, nothing typed, no submit and nothing in a form, no act the kit marks costly (`data-tone` danger/warning) and none whose own name is the write, in either language — and nothing is FOCUSED, because focus cannot be taken from one field without leaving another, and a field that saves itself when the reader leaves it saves on exactly that. Where a reading needs the focused state, as the focus-ring rule does, Chrome is asked to PAINT `:focus-visible` and release it again: the cascade answers, focus never moves and no event is dispatched. Each overlay closes with Escape, confirmed closed; a record is descended at most twice; the walk stops at twenty-four surfaces per screen and NAMES each door it left. Each surface is measured once no request is in flight, and the measurable probes of `@lotics/ui` docs/reviewing.md run over the DOM, each finding printing the rule it IS — its law, its section and the one edit that answers it — so the numbers need no key. What they exempt is what the screen itself declares: a register's ordinal gutter (a column counting to the row count is the shape's numbering, which no app can treat), a strip whose list carries `data-order="sequence"` (a lifecycle rail, which composition.md permits under a screen's tabs), a hairline or `clip-path`-clipped leaf (the visually-hidden node a control plants for a screen reader), and a leaf whose own computed line clamp states a count over a sentence. **A meter counts as an encoding only where it draws a POSITION** — `aria-valuenow` inside a range with room left. A bar pinned at its own maximum — what a meter alarmed AT its maximum draws on every alarmed row — reads the same as every value above it, and one with no maximum states none; both are counted in the census's `devices` and out of its `encoded`, so "nothing drawn" and "drawn and saying nothing" never read alike. The bare values that remain are grouped into columns, each named by the heading over it, so a finding says WHICH slot draws its figures as words. **Two finding classes read what geometry cannot.** `clutter` (docs/hierarchy.md): a second primary act, a value in two places on a record, a box reserving more lines than it holds, markers on over half a form's fields, a second accent. `right_form` (docs/screen.md's device index): a boolean as a two-option select, a day run as a repeated date column where the kit ships `Schedule`. Each prints a line per rule fired, with three offenders. **Four read room and counts** (reviewing.md §8l–§8o): `reserved_blank`, a register column 40px or wider that is blank (no text, no control) on more than half the rows in view, or a row-act gutter leaving 40px beside the acts on them — a row a complete register draws for nothing stored (`data-ghost`, a day nobody filed) is no row of the set; `shed_with_room`, a column the register's width gave up (the kit states them as `data-shed`) that, with the gap between two columns, fits in what its slots hold beyond what each must seat — its heading and widest ink, a fixed column's whole width, the acts most rows have; `action_off_heading`, a section's create (a control stating `data-create`) drawn in its body rather than its heading row; and `partial_count`, a count or `n of N` equal to the rows in hand of a read the server cut on the screen being walked, where the server's own count of the same alias, params and filter — asked by this CLI, a read — says otherwise. **Three read a record's anatomy** (docs/hierarchy.md), over each record on the surface with the title heading it: `record_primary`, a primary act in a block's heading row — a block's add is secondary beside the record's own acts (two under one heading are `clutter`'s); `fact_repeats`, a field whose value reads the same as the header's own words, whatever its case; and `empty_block`, a block that draws nothing under its heading, not even the line saying it is empty. A census line per screen (text runs, money strings, bare values against the devices reading, tab strips) prints first, so a clean verdict over a screen that rendered nothing cannot pass; a screen that renders no text is itself a finding, and one still changing after fifteen seconds is measured as it is. **A cold dependency optimisation is waited out, not measured**: the first paint has its own bound, far longer than settle's, since an app's modules load after its document completes and Vite holds them; only a MOUNTED, idle, textless frame is blank at once, and the finding names the wait and its blocker. It also RELOADS the page under the probe, so a width is measured again once; a second reload is a page that keeps moving and fails. **`--screen <label>` and `--width <n>` narrow a run** (repeatable, comma-separated; the label matches case-insensitively as a substring), for the author iterating on one screen who would otherwise pay a typecheck, a Vite boot and every screen at both widths on every edit; the clean verdict then names only the widths covered, and a `--screen` matching nothing is refused. **`--shots <dir>` writes what the run measured** — a `<surface>@<width>.png` and a `<surface>@<width>.json` per surface, off the SAME settled frame the probes read, so a shot and a finding can never describe different pixels. The PNG is the WHOLE surface: an app scrolls inside a box of its own, so the window is grown to the height the probe measured and put back, and an overlay a resize dismissed is shot as the viewport, the sidecar saying so (`app dev`'s header band is in it — the band the app was laid out under). The JSON is the half a picture cannot carry: the nav's first item's left edge, the title and first section heading, the primary acts by name, label/value pairs against what the folds state, values drawn twice, money that wraps, text the layout cut, and the console errors and uncaught exceptions the frame raised. **An uncaught exception is a finding** (`crash`, its first line, on the surface it was thrown reaching); a console error is not, since every React development warning is one. **A register reading the app states and a screen drew nowhere is printed under "Stated and not drawn:"** with the runtime's reason (narrowed by a chip, no part holding more than one row, no rows in view), and is not a finding. Named `<nn>-<slug>` plus one `__<step>` per door taken (`__record`, `__menu`, `__child`, `__fold`, `__dialog-<n>`, `__tab-<its name>`, `__section-<its name>`); `<nn>` is the walk's index, which lists the directory in reading order and keeps two labels that fold to one ASCII slug apart. The row a record surface opened from is the sidecar's `opened_from` and that screen's census line, never a file name — it is a person's data. The directory is created if missing, and refused before the dev server boots when it cannot be; `--shots` without `--screens` is refused. Findings exit 1 like the rest. **Where it renders**: the app's files are copied into the render project of its DEPENDENCY SET — `~/.lotics/render/<key>/`, keyed by the manifest's `dependencies` and `devDependencies` as written, a `file:` tarball by its bytes — the one `app preview` renders in, so the install and Vite's dependency optimisation are paid once per set rather than per app; one render holds a project at a time (`<key>.lock` — a second run waits, naming the app and pid, and takes over a dead hold). Where that install resolved another runtime or kit than the app's own, the run says which, since a deploy builds the app's own. The run ends with what each phase cost: `preflight · install|reuse · walk`. **`--changed`** walks only when something the screens READ has moved since this checkout's last walk — each part of `app.json` (every record apart), each declared query, each bundled source file, the field map, the ranges, the kit the render resolved and the CLI version, whose probes measure them — read against `.lotics/screens_check.json`, which every walk writes; with nothing moved, that walk's report is printed again under its timestamp and still fails the run. An app's screens all read its one spec, so a move walks the app whole. It walks, saying why, with no earlier walk, a different `--screen`/`--width`, or `LOTICS_UI_SRC` set (a linked working copy has no fingerprint); the workspace's rows are not an input. `--changed` without `--screens`, or beside `--shots`, is refused. |
+| `lotics app preview <model.json>#<app> [--shots <dir>] [--width <n>] [--screen <label>] [--kit <path>]` | **The app a model states, RENDERED — before a table exists, and with no credential in the process.** The model is checked offline, the named app is compiled against the workspace the model WOULD become (synthetic `tbl_`/`fld_`/`opt_`/`grp_` ids, minted positionally off the file), and every read the app makes is answered from the model's own `rows` — projected under the columns the query names, narrowed by the params it declares and sorted the way it states, so a child block under a record holds that record's rows and not every parent's. An entity the app reads that states no `rows` gets three synthesized, so a register never measures clean over nothing. Then the SAME headless walk `app check --screens` takes: the register, the record its first row opens in its door, and the add dialog, at 1280 and 375, measured by the same probes. `--shots <dir>` writes a PNG and a sidecar per surface into `<dir>/<app>/`, so apps previewed into one directory never overwrite each other. **Exits 1 on any finding** — an exception the app throws is one — exactly as the check does, and prints what each phase cost — bind, install-or-reuse, walk. **Nothing is created and nothing is reached** — no app row, no table, no workspace; the only network it needs is the npm registry, and only when the kit has moved. It renders in the cached render project of its dependency set (`~/.lotics/render/<key>/`, shared with `app check --screens` and with every model whose app lists the same set, and held by one render at a time — a second run waits, naming the app and pid holding it, and takes over a hold whose process is gone): the runtime and its kit are PUBLISHED packages, so a project is what a preview needs to install them into — the CLI has no renderer of its own. **It installs the runtime this CLI was built for**, and the kit through it, never `latest`: the spec this binary's generator writes is the one that runtime reads, and a version that published mid-session is a runtime nobody checked the model against. A version the registry does not serve yet is refused before npm runs, naming `--kit` — needed only until that runtime is published. **`--kit <checkout>`** (repeatable) — a checkout's ROOT, which names the app-runtime and the ui it holds, or one of those two packages — renders against that checkout instead — built, packed and proven the way `lotics app kit` installs one into an app — and the run prints that it rendered against a LOCAL CHECKOUT, first and last, so the render is never read as the published one. The project is a CACHE and never a source: `npm install` runs on the first render of a set and again only when a checkout's tarball did, and every file in it is rewritten from the model on each run. **A document a row names by path is served** from the project's own static root, as the named file a workspace would hold, so a files block and a record's picture read what the model states; a `fil_` id is carried by name. **`--record <ref>`** opens that row's record by its own address — a row of the app's entity, by the ref the model gives it, open or closed — where the walk otherwise opens the first row the register lists; a ref naming no such row is refused with the refs there are. **Every read is answered at the BRIDGE**, not inside the page. The app SDK's design-time fixture (`registerMockFixture` + `?__mock=1`) is the wrong half of this: the flag lives in the app's own url, and the first record page is a navigation the app's router performs — so from that surface onwards the flag is gone and every read falls through anyway. The bridge is the one place every call arrives whatever the url says, so `query`, `field_options`, `members` and `context` are answered there from one reading of the rows, and the generated entry ships EXACTLY as `app create --from` writes it. **A preview writes nothing**: a workflow, an upload or an agent run meets the same read gate `app check --screens` uses, so the app draws its own refusal path rather than reporting a save nothing moved for, and the call is reported above the census the way the check reports one. **What came out empty, it names**: a `formula`, `rollup`, `lookup` or `autonumber` is computed here rather than stated in the model, and the census lists each computed column that stayed empty on EVERY row of a table that has rows — named by outcome, since a total drawn blank is either the app's own answer or a hole and only the run can say which. The census reads the CELLS and not the field types, so nothing is exempt by kind. **A computed column that came out `#ERROR:` is named too, and the run refuses it** before anything is installed: a wall of red measures clean, and a run that drew it and exited 0 would tell its author the app is fine. **It computes in UTC**, because a model states no timezone — a row's `@today`, an autonumber's `{YEAR}` and a record's own clock all land on the run's own date at UTC midnight, so two runs of one model draw the same register wherever they are made. |
 | `lotics app workflow set <alias>` | Push the edited `src/workflows/<alias>.ts` body through `set_app_workflow` (the single author of `apps.workflows`). Reads the body from disk (header + `/// <reference>` + `export {};` marker + the `__workflow` wrapper all stripped) + the typed `inputs`/`outputs` **and the `description`** from `package.json#lotics.workflows.<alias>`; the **server** re-verifies the body and echoes the bound `outputs` (declared, else DERIVED from `return({ data })`). The `description` is the one line an agent reads when choosing between the app's aliases (the workflow counterpart to a query's) — authored in the manifest so it lives beside the body in version control and rides every push; omit it and the workflow keeps whatever description it already has, so a push can never blank one set elsewhere. When the manifest declared NO `outputs`, the DERIVED echo is written back into `package.json#lotics.workflows.<alias>.outputs` (a SURGICAL write — preserves `knowledge`/`config` and every other manifest field) and that alias's types are refreshed in place, so `useWorkflow("<alias>")`'s `result.data` is typed immediately with no hand-copy and no second `lotics app codegen`; an explicitly-declared `outputs` is authoritative and never overwritten. A deploy runs this same verb for every alias whose declaration or body is ahead of the app, so this command is the one-alias spelling of what a release does, not a step a release leaves to a person. Clear error + non-zero exit on a missing file, an alias absent from the manifest, or a verify failure. A push also prints any non-blocking verify warnings, including an input the alias declares that the body never reads. A first bind MINTS the workflow row, and the id it echoes is written back into `package.json#lotics.workflows.<alias>.workflow_id` — the same surgical write the derived `outputs` gets. Without it a hand-declared alias ended up shaped unlike its siblings, so anything reading the manifest (an audit, a port to another workspace, a person comparing two blocks) had to treat a missing id as normal, which is exactly how a genuinely missing one stops being visible. **`--acknowledge-breaking-api`** carries this write out even though it breaks what the app's published API promises, snapshotting the broken contract as a new version; without it such a write is refused and every breaking change is named (see `app api`). |
 | `lotics app agent set <alias> [--acknowledge-breaking-api]` | Push `src/agents/<alias>.md` — plus `inputs`/`outputs` when `package.json#lotics.agents.<alias>` declares them — through `set_app_agent`. The agent mirror of `app workflow set`, and the deploy-free authoring path for an agent's prose and its typed edges. **It sends only those fields.** Everything else is absent, and absent means unchanged, so a declaration this CLI does not model cannot be reverted by a push from a checkout that predates it — the chat authoring agent's `knowledge_doc_ids`, another operator's `query_aliases` grant. To change one of those, call `set_app_agent` with just that field (`lotics run set_app_agent '{"app_id":…,"alias":…,"tool_names":[…]}'` — it merges), then `app pull` to bring the manifest back in step. **CREATES the alias when the app has not bound one yet**, so a new agent is authored the same way a new workflow is: write the prose, declare the typed half, push. A create needs the prose file (an agent without instructions is not an agent); it is gated on nothing else, because what keeps a binding alive is a `useAppAgentRun("<alias>")` call site in the shipped bundle — a deploy prunes an agent the bundle never names, manifest entry or not. The prose push is a conditional write against the fingerprint this project last saw, so it is refused rather than allowed to overwrite prose someone else changed. Clear error + non-zero exit when there is no prose file and nothing declared to push instead, when a create has no prose to create from, or when the file is empty once the header is stripped. **`--acknowledge-breaking-api`** carries this write out even though it breaks what the app's published API promises, snapshotting the broken contract as a new version: an agent's alias and its declared `inputs`/`outputs` are in that contract too, so renaming one or changing what it accepts breaks a caller nobody here can redeploy. Without the flag such a write is refused and every breaking change is named (see `app api`). |
 | `lotics app query set <alias>` \| `--all` | Push `package.json#lotics.queries` (`{ ast, params? }` per alias) to `apps.queries` through `set_app_query` — **the only author of a query binding**, the mirror of `app workflow set`. A deploy pushes a DRIFTED declaration through this same verb before it ships (see `app deploy`), so this is the explicit single-alias path, not the only way a query reaches the app. The **server** validates each one exactly as it always did (alias identifier, workspace-only tables, resolvable fields, declared params). `--all` pushes every declared alias, alias-sorted — after holding every one to the server's own query gate and pushing none when any is refused; a refusal past that stops at the first failure and names what already landed. Clear error + non-zero exit on an alias absent from the manifest or a validation failure. **The manifest is the WHOLE declaration**: `set_app_query` merges, so the push sends every optional key, `null` for one the manifest no longer carries. After the push it regenerates `.lotics/app_queries.d.ts` from the manifest, so the types the next `npm run typecheck` reads match what was just pushed. **`--acknowledge-breaking-api`** carries this write out even though it breaks what the app's published API promises, snapshotting the broken contract as a new version; without it such a write is refused and every breaking change is named (see `app api`). |

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@lotics/cli",
-  "version": "0.258.0",
+  "version": "0.259.0",
   "description": "Lotics SDK and CLI for AI agents",
   "type": "module",
   "bin": {