@pipelex/sdk 0.26.0 → 0.28.0

package/docs/architecture.md

@@ -72,7 +72,7 @@ A run request carries its method inline exactly one of three ways: `mthds_conten
 
 - **It is a named option, not an `extra` entry.** Typing its own platform's arguments is the one job a layer-3 client exists to do; `extra` stays the escape hatch for extensions this client does not know about (a third-party server, or a newer version of ours). That split, and the reserved-keys guard that enforces it, are normative for every client in the layered stack: a client types the arguments of the platform it is built for, and only the layer that owns an argument may validate or reject it. It was previously passed as `extra: { method_id }`, which worked but documented nothing and validated nothing.
 - **The guard is per layer.** `method_id` is reserved on `extra` _here_ and must never become reserved in `mthds` / `mthds-python`: a protocol client talking to another vendor's server has no business rejecting that vendor's arguments.
-- **Pure pass-through — nothing is expanded client-side.** The platform resolves the id against the org's catalog. The same posture holds across the tooling routes: `validate` / `resolve` / `codegen` take a typed `method_id` the platform resolves server-side (see the surface list below), and so does `prepareInputs`, which composes a `validate` of its own rather than a route call. `/v1/build/*` is the last family with no by-id form — it takes `files` only, and a stored method is expanded explicitly with `getMethodClosure`, which stays public as the local expansion utility.
+- **Pure pass-through — nothing is expanded client-side.** The platform resolves the id against the org's catalog. The same posture holds across the tooling routes: `validate` / `resolve` / `codegen` / `pipeIo` take a typed `method_id` the platform resolves server-side (see the surface list below), and so does `prepareInputs`, which composes a `pipeIo` of its own. `/v1/build/*` is the last family with no by-id form — it takes `files` only, and a stored method is expanded explicitly with `getMethodClosure`, which stays public as the local expansion utility.
 - **Alone it is a run source; alongside an inline source it is linkage.** With no inline source the platform resolves and runs the stored method. With one, the inline source is what RUNS (precedence) and the id is recorded as run-history linkage on the Run row — the index key `GET /v1/runs?method_id=` queries. Rejecting the combination is not an option: it once orphaned every unsaved-bundle run from its method. The client therefore counts `method_id` towards its "something to run" precondition, and never treats it as an exclusivity violation.
 - **It rides the blocking fallback too.** `startAndWaitForResult` degrades to `POST /v1/execute` against a runner with no run store, and the selector goes with it — so a bare `pipelex-api` answers a `422` that names the key (its Rule 4 diagnosis: "a hosted-only selector must be sent to the hosted API") instead of the client silently dropping it and running something else.
 - **An empty string is treated as absent.** `method_id: ""` selects no method and links nothing, so it is not sent and does not satisfy the precondition.
@@ -84,12 +84,12 @@ Flat `src/` (mirrors `mthds-js`'s `runners/api` flatness — the SDK has one cli
 - `client.ts` — `PipelexApiClient`: the `request()` pipeline (auth, base URL, timeouts/abort, problem-details parsing) and every route. `implements MTHDSProtocol<DictPipeOutput>` so the protocol-execution methods stay shaped like the standard.
 - `models.ts` — Dict-serialized concretes (`DictStuff` / `DictWorkingMemory` / `DictPipeOutput` / `DictRunResultExecute`), the `/v1/validate` surface (`PipelexValidationResult`, `ValidationErrorItem`, and the `SuggestedFix` / `FixOp` repair vocabulary its items carry — the standard's declarations, restated because `mthds` exports them only from its package root and pinned to it by a type-level test, see [errors.md](./errors.md#a-validation-item--validationerroritem)), the tools surface (`Diagnostic` / `LintResponse` / `FormatResponse`), and the crate-family request/response models — the shared `CrateRequestBase` / `CrateInvalidReport` / `GeneratedArtifact` vocabulary plus the `/v1/build/*`, `/v1/resolve` and `/v1/codegen` shapes built on it.
 - `product-models.ts` — the snake_case wire shapes for the hosted-product management routes (profile, methods catalog, organizations, billing, API keys, onboarding, storage, runs list/detail/update). Each interface models only the fields the product actually consumes — no speculative mirrors.
-- `error-models.ts` — the error-report wire models: `RunErrorReport`, the runner's `ErrorReport` typed with every field it carries (with `UserAction`, `ProviderErrorMetadata` and `MigrationErrorBlock`), which is the one type for a failed run's report on `RunPublic.error`, `PipelineRun.error`, the failed arm of `RunResultState` and `RunFailedError.error`; and `ProblemDetails` / `FieldError`, the typed members of a problem document that `ApiResponseError` exposes. Every field is optional and every interface keeps an index signature, because the runner owns the shape and adds fields without asking. Types only: it imports the `ValidationErrorItem` type and nothing else. See [errors.md](./errors.md).
-- `method-source.ts` — `methodSourceToContents`, the canonical parser turning a stored method's polymorphic `mthds` source (raw single-bundle text XOR a JSON `{name, content}[]` file array) into a flat list of file contents. A verbatim port of the platform's `_method_source_to_contents` so the SDK and runtime read a stored source identically; exported from the barrel and consumed by `getMethodClosure`. See [input-preparation.md](./input-preparation.md#getmethodclosure--the-explicit-expansion-utility).
+- `error-models.ts` — the error-report wire models: `RunErrorReport`, the runner's `ErrorReport` typed with every field it carries (with `UserAction`, `ProviderErrorMetadata` and `MigrationErrorBlock`), which is the one type for a failed run's report on `RunPublic.error`, `PipelineRun.error`, `RunHistoryItem.error`, the failed arm of `RunResultState` and `RunFailedError.error`; and `ProblemDetails` / `FieldError`, the typed members of a problem document that `ApiResponseError` exposes. Every field is optional and every interface keeps an index signature, because the runner owns the shape and adds fields without asking. Types only: it imports the `ValidationErrorItem` type and nothing else. See [errors.md](./errors.md).
+- `method-source.ts` — `methodSourceToContents`, the canonical parser turning a stored method's polymorphic `mthds` source (raw single-bundle text XOR a JSON `{name, content}[]` file array) into a flat list of file contents. A verbatim port of the platform's `method_source_to_contents` so the SDK and runtime read a stored source identically; exported from the barrel and consumed by `getMethodClosure`. See [input-preparation.md](./input-preparation.md#getmethodclosure--the-explicit-expansion-utility).
 - `codegen-check.ts` — `runCodegenCheck`, the **pure** offline drift check over a committed codegen tree: the stamp grammar, the `codegen.lock` parser, the hash, and the drift taxonomy, ported from pipelex's `codegen/check.py` so a verdict computed here equals the one `pipelex codegen check` computes over the same bytes (the drift `detail` sentences included, verbatim). It touches neither the client nor `mthds/protocol` — its only import is `smol-toml` — and it hashes through WebCrypto rather than `node:crypto` so it adds no Node builtin to the barrel's import graph (`upload.ts` still names `node:fs/promises`, so a browser-targeting bundler must mark `node:*` external). It raises `CodegenLockError`, which extends `Error` rather than the protocol base. No filesystem, no fetch, no API key — the caller walks its own tree and hands in the text. See [crate-routes.md](./crate-routes.md#the-offline-check--runcodegencheck).
 - `upload-grant.ts` — `uploadWithGrant`, the sender of an upload grant (`requestUploadGrant`, `POST /v1/upload/grant`): the file as the raw body of a `PUT` to the grant's URL with its signed headers unchanged, redirects refused, the whole exchange bounded by a time limit scaled by the file's size unless the caller passes `timeoutMs`, only the first 16 KiB of storage's error body read, and storage's refusals mapped onto the input-preparation errors (`RejectedAssetError` with a `code` for a `4xx` such as the `412` of a used grant or the `403` of a signature mismatch, `UploadTransportError` with a `code` for the time limit running out, a `5xx`, storage's `RequestTimeout`, a `409` conflict, a redirect or no response). It is the browser-safe `@pipelex/sdk/upload` entry, so it imports the error classes, the timer bound in `timers.ts` and the wire types only, never the client. See [input-preparation.md](./input-preparation.md#uploading-with-a-grant--requestuploadgrant--uploadwithgrant).
 - `artifacts.ts` — the artifact stack, the download twin of `prepareInputs`, in layers: `locateArtifacts` (a pure, exported walk for `pipelex-storage://` references, each with every `$`-rooted path it sits at) and `collectArtifacts` (the same walk's bare references), `resolveArtifacts` (the bulk resolve route over a whole list, chunked at its bound), `fetchArtifact` (one reference as a bounded `Response`) and `downloadArtifacts` (Node-only, a run's produced files saved under a directory as a produced verdict), plus `artifactFilename`, the rule that names a saved file after the field it fills. The three network operations are client methods over a small structural seam (`ArtifactCapableClient`: the raw `resolveStorageUrls` call and `getRunResult`); the module's static import graph stays free of Node builtins, because `node:fs/promises`, `node:path` and `undici` are loaded by dynamic import behind `isNodeRuntime()` — a browser-targeting bundler still has to mark `node:*` and `undici` external, as it already must for `upload.ts`. `undici` is a runtime dependency for one reason: the fetch runs on an `Agent` carrying `headersTimeout` and `bodyTimeout`, the per-stall bounds an `AbortSignal` budget does not give, and Node exposes its bundled undici's `fetch` but not its `Agent`. See [artifact-download.md](./artifact-download.md).
-- `runs.ts` — the run-lifecycle types (`RunStatus`, `RunRead`, `RunResults`, `RunResultState`) and the single poll loop (`pollUntilResult`). `RunResults.main_stuff` (the resolved main output content) is always present for a completed run: on the hosted path it is the `main_stuff.json` S3 artifact; on the bare-runner blocking path the SDK resolves it from the returned working memory via the response's `main_stuff_name`, so both paths deliver the same shape. Consumers read `main_stuff` directly. `working_memory` — every named stuff of the run, `{ root, aliases }`, typed `DictWorkingMemory` from `models.ts` — reads the same on both paths: the hosted path relays the `working_memory.json` artifact, and the blocking path lifts it off `pipe_output`, which still rides whole (blocking path only). `graph_spec` — the executed graph — also reaches the client on both paths: the hosted path relays the `graphspec.json` artifact, and the blocking path lifts it off `pipe_output` beside `graph_assembly_error`, the graph's twin of `usage_assembly_error`. That error field is declared ahead of the wire on the hosted side, where the platform's results body relays no such key and the field is therefore absent, reading `undefined` rather than `null`. A completed run that cannot deliver a main stuff throws `MissingMainStuffError`. `runs.ts` also owns `TokensUsageRecord` — one client-facing usage record per inference call, carried by `RunResults.tokens_usages` on both paths. That is a mirror of a wire contract this SDK does not own (inference accounting is a Pipelex runtime extension rather than part of the MTHDS standard): every field is optional and the interface carries an index signature so pre-contract artifacts (relayed verbatim, never migrated) still type-check. See [`run-usage.md`](./run-usage.md) for the field reference, the cost/null semantics, and the old-artifact rules, and [`run-results.md`](./run-results.md) for the field-by-field account of `RunResults` — the durable handle, the output and how to read it, the run's working memory, the executed graph and how to render or keep it, and the produced files with their short-lived links.
+- `runs.ts` — the run-lifecycle types (`RunStatus`, `RunRead`, `RunResults`, `RunResultState`) and the single poll loop (`pollUntilResult`). `RunResults.main_stuff` (the resolved main output content) is always present for a completed run: on the hosted path it is the `main_stuff.json` S3 artifact; on the bare-runner blocking path the SDK resolves it from the returned working memory via the response's `main_stuff_name`, so both paths deliver the same shape. Consumers read `main_stuff` directly. `working_memory` — every named stuff of the run, `{ root, aliases }`, typed `DictWorkingMemory` from `models.ts` — reads the same on both paths: the hosted path relays the `working_memory.json` artifact, and the blocking path lifts it off `pipe_output`, which still rides whole (blocking path only). `graph_spec` — the executed graph — also reaches the client on both paths: the hosted path relays the `graphspec.json` artifact, and the blocking path lifts it off `pipe_output` beside `graph_assembly_error`, the graph's twin of `usage_assembly_error`. That error field is declared ahead of the wire on the hosted side, where the platform's results body relays no such key and the field is therefore absent, reading `undefined` rather than `null`. A completed run that cannot deliver a main stuff throws `MissingMainStuffError` when the read asked for it. `runs.ts` also declares `RUN_RESULT_ARTIFACTS` / `RunResultArtifact`, the artifacts a results read can be narrowed to with `getRunResult`'s `artifacts` option; with a selection an unselected artifact is absent while a selected unwritten one is `null`. `runs.ts` also owns `TokensUsageRecord` — one client-facing usage record per inference call, carried by `RunResults.tokens_usages` on both paths. That is a mirror of a wire contract this SDK does not own (inference accounting is a Pipelex runtime extension rather than part of the MTHDS standard): every field is optional and the interface carries an index signature so pre-contract artifacts (relayed verbatim, never migrated) still type-check. See [`run-usage.md`](./run-usage.md) for the field reference, the cost/null semantics, and the old-artifact rules, and [`run-results.md`](./run-results.md) for the field-by-field account of `RunResults` — the durable handle, the output and how to read it, the run's working memory, the executed graph and how to render or keep it, and the produced files with their short-lived links.
 - `usage.ts` — `summarizeUsage`, the **pure** fold of a run's usage pair (`tokens_usages` / `usage_assembly_error`) into one `UsageSummary`: a `state` (`records`, `no_inference` or `unavailable`), the null-aware total cost with its `cost_partial` flag, the `input` and `output` token totals, the call count, the relayed assembly error, and a `by_pipe` rollup sorted by cost with the unattributed calls grouped under a `null` `pipe_code`. It applies the rules [`run-usage.md`](./run-usage.md#summarizing-a-run--summarizeusage) states, so that every consumer reads one computation rather than re-deriving those rules by hand. It imports types from `runs.ts` and nothing else: no client, no fetch, no `mthds/protocol`.
 - `execute-result.ts` — `PipelexExecuteResult`, a `DictRunResultExecute` subtype that adds a resolved `.main_stuff` accessor (dug out of `pipe_output`'s working memory via the response's `main_stuff_name`, throwing `MissingMainStuffError` if unlocatable). `execute()` returns it, giving blocking and durable results the **same output accessor** — `result.main_stuff` — so callers never branch on which path ran. The neutral `mthds` wire model leaves `main_stuff_name` in its extension index signature; this Pipelex-branded subtype declares it as a typed field (Pipelex owns that concept). `resultsFromExecute` (the public lift from that result onto `RunResults`, in the same module) reads that accessor too, keeping the resolution single-sourced; the `executeBlocking` fallback is its only caller inside the SDK, and a consumer driving `execute()` directly calls it by name.
 - `errors.ts` — the transport and run-lifecycle typed errors, the input-preparation family under `InputPreparationError`, and the artifact family under `ArtifactOperationError` (`ScopeUnavailableError`, `ArtifactFetchError`, `ArtifactAuthenticationError`, the last carrying the download verdict so far); all derive from the protocol-base `PipelineRequestError` (re-exported from `mthds/protocol`) except `ClientAuthenticationError`. `ApiResponseError` carries every member of the problem document it was built from — the branch fields `errorDomain` and `type`, `retryable`, `userAction`, `requestId` (the body's, else the `X-Request-ID` header's), the native `code` and `errorType`, the platform's `errors[]`, the inference members, and the decoded document whole as `problemDocument`; its member names and types match `mthds`'s own `ApiResponseError` wherever the two overlap, which a type-level test pins because this package may not import `mthds/errors`. `RunFailedError` carries the run's `status` and its stored report as `error`. It is not the whole error surface: `codegen-check.ts` exports `CodegenLockError`, which extends plain `Error` because nothing was requested over the wire.
@@ -101,14 +101,14 @@ Flat `src/` (mirrors `mthds-js`'s `runners/api` flatness — the SDK has one cli
 ## Client surface (current)
 
 - **Protocol execution:** `execute` (returns a `PipelexExecuteResult` with a resolved `.main_stuff`), `start`, `validate` (returns `PipelexValidationResult`), `validateFiles`, `models`, `version`.
-- **Two independent opt-in lists on `validate`.** `render` asks the server to render a *view format* and adds a `rendered_<format>` field; `views` asks for a *structured view* and adds a same-named top-level field — today only `input_form`, the per-pipe input-form descriptors. Each resolves its own tokens against its own supported set and unknown tokens are lenient-ignored on both, never a `422`. The client always asks for `render: ["markdown"]` but sends `views` only when a caller names a token, because an opt-in view exists precisely so the default response stays byte-identical for the consumers that discard it. `pipelex-api` gates `input_form` on the token as of 0.18.0; a 0.17.0 runner resolved no token and emitted the field on every valid verdict, which is why `PipelexValidationReport.input_form` is typed optional rather than required — the one typing that reads a response from either runner correctly.
+- **Two independent opt-in lists on `validate`.** `render` asks the server to render a *view format* and adds a `rendered_<format>` field; `views` asks for a *structured view* and adds a same-named top-level field — today `input_form` and `output_form`, the per-pipe form descriptors. A caller that wants the forms without the dry-run verdict reads `pipeIo` instead, which derives them with no dry run. Each resolves its own tokens against its own supported set and unknown tokens are lenient-ignored on both, never a `422`. The client always asks for `render: ["markdown"]` but sends `views` only when a caller names a token, because an opt-in view exists precisely so the default response stays byte-identical for the consumers that discard it. `pipelex-api` gates `input_form` on the token as of 0.18.0; a 0.17.0 runner resolved no token and emitted the field on every valid verdict, which is why `PipelexValidationReport.input_form` is typed optional rather than required — the one typing that reads a response from either runner correctly.
 - **Method-bundle transport (pipelex-api extension):** `execute` / `start` / the blocking fallback accept the whole method — the `.mthds` plus its `funcs/*.py`, `structures/*.py`, `requirements.txt` — instead of only the inline `mthds_contents` text, so custom PipeFunc Python travels with the method. Two equivalent, mutually exclusive encodings: `files` (a `{ relativePath: text }` map) and `bundle_b64` (the same bundle as a base64 zip). A bundle is self-contained, so it satisfies the "something to run" precondition on its own — neither `pipe_code` nor `mthds_contents` is required beside it. See [run-sources](#run-sources) above.
 - **Method selectors on the run methods:** `execute` / `start` / `startAndWaitForResult` take the Pipelex API's `method_ref` (an address the runner resolves; provenance comes back typed as `method_provenance`) and the platform's `method_id` as **typed named options** (`PipelexRunOptions` / `PipelexStartOptions`), never `extra` entries. See [Pipelex API run extensions](#pipelex-api-run-extensions-method_ref) and [hosted run extensions](#hosted-run-extensions-method_id) above.
 - **Method selectors on `validate`:** the first argument is inline contents (`string[]`) or a `ValidateMethodSelector` — `{ method_ref }` (runner-resolved by address) or `{ method_id }` (hosted, platform-resolved) — under the strict tooling XOR: exactly one selector, no linkage exception, `mthds_sources` legal only beside inline contents. A selector-resolution failure is a non-2xx `ApiResponseError`, never an `is_valid: false` verdict.
-- **Build helpers (`/v1/build/*`):** `buildInputs`, `buildOutput`, `buildRunner`, `concept`, `pipeSpec`. The three projections share one closure envelope (`files: [{content, source?}]` XOR a `method_ref` — the address form server-resolved, the registry form still `501`) plus an optional **qualified** `pipe_ref` that defaults to the closure's `main_pipe`, and each returns a **discriminated 200 verdict** (`is_valid`) rather than a bare payload — same discipline as `validate`. The build routes take **no `method_id`** (`method_id?: never` on `BuildRequestBase`): the hosted tooling selector deliberately excludes them, so a stored method is expanded first with `getMethodClosure`. **Nothing inside this SDK calls them** — `prepareInputs` reads its signature from the input-form descriptor on the validate report — so they exist for external consumers only, and are retired with the rest of `/v1/build/*` once those consumers project the template client-side. See [build-routes.md](./build-routes.md).
-- **Crate routes (`/v1/resolve`, `/v1/codegen`):** `resolve` emits the **normalized library crate** (the MTHDS standard's Library Crate Format, fingerprint riding inside the payload); `codegen` projects that crate through two explicit axes — `kind` (`types`) × `target` (`ts-zod` / `python-pydantic` / `python-structures`) — into stamped artifacts plus their `codegen.lock`. They share the build routes' closure envelope and its **discriminated 200 verdict** (`CrateInvalidReport` is literally the same invalid arm), and additionally take the hosted `method_id` (`PipelexHostedToolingExtensions`) as a third closure form under the strict three-way XOR — a pass-through the platform resolves, meaningless against a bare runner. Writing the artifacts and lock verbatim reproduces a local `pipelex codegen types` run byte-for-byte, so the offline `pipelex codegen check` passes — the SDK stays transport-only and does not write files. Verifying such a tree afterwards is `runCodegenCheck`, a standalone pure helper rather than a client method, because it needs no server at all. See [crate-routes.md](./crate-routes.md).
+- **Build helpers (`/v1/build/*`):** `buildInputs`, `buildOutput`, `buildRunner`, `concept`, `pipeSpec`. The three projections share one closure envelope (`files: [{content, source?}]` XOR a `method_ref` — the address form server-resolved, the registry form still `501`) plus an optional **qualified** `pipe_ref` that defaults to the closure's `main_pipe`, and each returns a **discriminated 200 verdict** (`is_valid`) rather than a bare payload — same discipline as `validate`. The build routes take **no `method_id`** (`method_id?: never` on `BuildRequestBase`): the hosted tooling selector deliberately excludes them, so a stored method is expanded first with `getMethodClosure`. **Nothing inside this SDK calls them** — `prepareInputs` reads its signature from the input-form descriptor `pipeIo` returns — so they exist for external consumers only, and are retired with the rest of `/v1/build/*` once those consumers project the template client-side. See [build-routes.md](./build-routes.md).
+- **Crate routes (`/v1/resolve`, `/v1/codegen`, `/v1/pipe-io`):** `resolve` emits the **normalized library crate** (the MTHDS standard's Library Crate Format, fingerprint riding inside the payload); `codegen` projects that crate through two explicit axes — `kind` (`types`) × `target` (`ts-zod` / `python-pydantic` / `python-structures`) — into stamped artifacts plus their `codegen.lock`. They share the build routes' closure envelope and its **discriminated 200 verdict** (`CrateInvalidReport` is literally the same invalid arm), and additionally take the hosted `method_id` (`PipelexHostedToolingExtensions`) as a third closure form under the strict three-way XOR — a pass-through the platform resolves, meaningless against a bare runner. Writing the artifacts and lock verbatim reproduces a local `pipelex codegen types` run byte-for-byte, so the offline `pipelex codegen check` passes — the SDK stays transport-only and does not write files. Verifying such a tree afterwards is `runCodegenCheck`, a standalone pure helper rather than a client method, because it needs no server at all. `pipeIo` derives a method's three I/O artifacts — the standard's `PipeIOContracts`, `InputForm` and `OutputForm`, imported from `mthds/protocol` — for the pipe the route selects (the request's qualified `pipe_ref`, else the method's own entry pipe) or for every pipe under `all_pipes`, with the runnability facts beside them and the closure's files echoed on `include_files`, and it runs no dry run; a selection it refuses is a `422` typed `EntryPipeNotFoundError` or `EntryPipeAmbiguousError`. See [crate-routes.md](./crate-routes.md).
 - **Tools (`/v1/lint`, `/v1/format`):** `lint`, `format` — single-file static diagnostics and canonical formatting, served by any pipelex-api runner (no inference, no bundle load). Both are **diagnostic endpoints** like `validate`: a malformed `.mthds` file is a produced verdict on a **200** carrying `diagnostics[]`, never a thrown error — `format` additionally echoes the content back unchanged with `changed: false`. Only a no-verdict condition is non-2xx: a malformed body, **malformed formatter `options`** (e.g. a non-numeric `column_width` ⇒ 422), auth, or a server fault, all mapped to the typed `ApiResponseError`. `lint` is the cheap static check (syntax / schema / semantic); it does not replace `validate`, which loads the bundle, resolves across files, and dry-runs the pipes.
-- **Durable run lifecycle (hosted extension — NOT protocol):** `getRunStatus`, `getRunResult`, `waitForResult`, `startAndWaitForResult` (handshakes `/v1/version`, takes the durable start+poll path on a hosted deployment, and self-heals to the blocking `execute` against a bare runner). A run that ended without completing answers the results read with a `409` whose `run_status` and `error` members carry its status and its stored error report; `getRunResult` turns it into the `failed` arm (`status`, `message` from `detail`, `error`) and the waiting calls on the durable path into a `RunFailedError` carrying the same (the blocking fallback's failure is the runner's `ApiResponseError`), the status read from the member, or from the `detail` sentence on a platform that does not send it yet. The status read serves the same report as `RunRead.error`. See [errors.md](./errors.md#a-failed-run--runerrorreport). Reading what a completed run consumed is `summarizeUsage(results)`, a standalone pure helper rather than a client method, because it folds the usage pair the results already carry and needs no server. See [run-usage.md](./run-usage.md#summarizing-a-run--summarizeusage).
+- **Durable run lifecycle (hosted extension — NOT protocol):** `getRunStatus`, `getRunResult` (with an optional `artifacts` selection sent as `?artifacts=`, see [run-results.md](./run-results.md#reading-only-some-artifacts--artifacts)), `waitForResult`, `startAndWaitForResult` (handshakes `/v1/version`, takes the durable start+poll path on a hosted deployment, and self-heals to the blocking `execute` against a bare runner). A run that ended without completing answers the results read with a `409` whose `run_status` and `error` members carry its status and its stored error report; `getRunResult` turns it into the `failed` arm (`status`, `message` from `detail`, `error`) and the waiting calls on the durable path into a `RunFailedError` carrying the same (the blocking fallback's failure is the runner's `ApiResponseError`), the status read from the member, or from the `detail` sentence on a platform that does not send it yet. The status read serves the same report as `RunRead.error`. See [errors.md](./errors.md#a-failed-run--runerrorreport). Reading what a completed run consumed is `summarizeUsage(results)`, a standalone pure helper rather than a client method, because it folds the usage pair the results already carry and needs no server. See [run-usage.md](./run-usage.md#summarizing-a-run--summarizeusage).
 - **Health:** `health` (origin-level `/health`).
 
 ### Product routes (hosted management surface)
@@ -121,8 +121,8 @@ The hosted catalog/account routes the webapp drives. All ride the same `{base}/v
 - **Billing:** `getSubscription`, `listPlans`, `listInvoices`, `createCheckout`, `changePlan` (409 `conflict` ⇒ no subscription), `getBillingPortal` (409 `conflict` ⇒ no subscription).
 - **Pipelex API keys (`plx_sk_…`):** `listPipelexApiKeys`, `createPipelexApiKey` (409 `pipelex_api_key_limit_reached`), `revokePipelexApiKey`, `rotatePipelexApiKey`. The plaintext `api_key` is returned once, on create/rotate.
 - **Onboarding:** `submitOnboarding`.
-- **Storage:** `resolveStorageUrl`, `resolveStorageUrls` (`POST /v1/resolve-storage-url/bulk` — the single route applied to a list of at most 100 references, one verdict per reference in request order, a refused reference a value on its item; a deployment without the route answers a `404` `ApiResponseError`), `upload` (JSON base64 — the multipart hop is browser→BFF only), `requestUploadGrant` (`POST /v1/upload/grant` — a presigned, create-only `PUT` for one new object, which the holder of the bytes sends with the standalone `uploadWithGrant`, also exported from the browser-safe `@pipelex/sdk/upload` entry; a declared size over the cap is a `413` `ApiResponseError`). The higher-level `uploadFile` / `prepareInputs` preparation surface is built on top of `upload`; `prepareInputs` additionally composes one `validate` call with `views: ["input_form"]` for the signature it walks, and takes the method as inline `files`, a `method_ref` or a `method_id` like every other method-taking operation — see [input-preparation.md](./input-preparation.md). The download twin — `locateArtifacts`, `collectArtifacts`, `resolveArtifacts`, `fetchArtifact`, `downloadArtifacts` — is built on top of `resolveStorageUrls` and `getRunResult`, never on the embedded `public_url`, and is documented on [artifact-download.md](./artifact-download.md); downloading is a separate explicit gesture, and no run call takes a `download` option.
-- **Runs list/detail/update:** `listRuns` (by `method_id`) returns **one page**, not the whole history — a `RunPage` (`{items, nextCursor}`) served from a time-ordered index, so its cost is the page size rather than the size of the history. `iterateRuns` is the async iterator that follows the cursor for callers who genuinely want everything (`for await (const run of client.iterateRuns(id))`); it is deliberately not a `listAllRuns(): Promise<PipelineRun[]>`, because an all-at-once helper needs a page cap and a cap means silently returning a truncated list. Both accept `createdFrom` / `createdTo`, applied server-side as index key conditions — these are **instants, not days** (ISO-8601 with a UTC offset; a naive timestamp is a 400), because only the caller knows which timezone's day it means. `getRunDetail` (`GET /v1/runs/{id}`) is the only call returning `mthds_contents` — what the run actually executed — and `inputs`; it is kept off the status read that pollers hit every few seconds and off the list, where the bundle would be multiplied by the page size. `updateRun` is the admin/manual status patch. The status/results/start lifecycle routes live in the run-lifecycle section above.
+- **Storage:** `resolveStorageUrl`, `resolveStorageUrls` (`POST /v1/resolve-storage-url/bulk` — the single route applied to a list of at most 100 references, one verdict per reference in request order, a refused reference a value on its item; a deployment without the route answers a `404` `ApiResponseError`), `upload` (JSON base64 — the multipart hop is browser→BFF only), `requestUploadGrant` (`POST /v1/upload/grant` — a presigned, create-only `PUT` for one new object, which the holder of the bytes sends with the standalone `uploadWithGrant`, also exported from the browser-safe `@pipelex/sdk/upload` entry; a declared size over the cap is a `413` `ApiResponseError`). The higher-level `uploadFile` / `prepareInputs` preparation surface is built on top of `upload`; `prepareInputs` additionally composes one `pipeIo` call, which selects the pipe and answers with the input-form descriptor it walks, and takes the method as inline `files`, a `method_ref` or a `method_id` like every other method-taking operation — see [input-preparation.md](./input-preparation.md). The download twin — `locateArtifacts`, `collectArtifacts`, `resolveArtifacts`, `fetchArtifact`, `downloadArtifacts` — is built on top of `resolveStorageUrls` and `getRunResult`, never on the embedded `public_url`, and is documented on [artifact-download.md](./artifact-download.md); downloading is a separate explicit gesture, and no run call takes a `download` option.
+- **Runs list/detail/update:** `listRuns` (by `method_id`) returns **one page**, not the whole history — a `RunPage` (`{items, nextCursor}`) served from a time-ordered index, so its cost is the page size rather than the size of the history. Its rows are `RunHistoryItem`, **not** `PipelineRun`: the id, `status`, `created_at`, `finished_at`, `pipe_code` and, for a failed run, `error` — what a history row shows. The run's organization, creator, method, workflow id and result location are not on the list; `getRunDetail` returns them. `iterateRuns` is the async iterator that follows the cursor for callers who genuinely want everything (`for await (const run of client.iterateRuns(id))`); it is deliberately not a `listAllRuns(): Promise<RunHistoryItem[]>`, because an all-at-once helper needs a page cap and a cap means silently returning a truncated list. Both accept `createdFrom` / `createdTo`, applied server-side as index key conditions — these are **instants, not days** (ISO-8601 with a UTC offset; a naive timestamp is a 400), because only the caller knows which timezone's day it means. `getRunDetail` (`GET /v1/runs/{id}`) is the only call returning `mthds_contents` — what the run actually executed — and `inputs`; it is kept off the status read that pollers hit every few seconds and off the list, where the bundle would be multiplied by the page size. `updateRun` is the admin/manual status patch. The status/results/start lifecycle routes live in the run-lifecycle section above.
 
 ## Hook bundle (`dist-hooks/check.mjs`)
 
@@ -133,4 +133,4 @@ The hosted catalog/account routes the webapp drives. All ride the same `{base}/v
 Two layers, split by whether a server is in the loop:
 
 - **Unit suites (`tests/*.test.ts`, `make test`):** mock the fetch boundary (`vi.spyOn(globalThis, "fetch")`) and pin the client's wire behavior — URLs, request bodies, status→error mapping, poll semantics. These run everywhere, including CI. The codegen-check suite is the exception that proves the rule: it has no server to mock, so it runs against **vendored real codegen output** under `tests/fixtures/codegen/` — artifacts and lock emitted by an actual `pipelex codegen types` run, committed beside the `.mthds` they came from. A hand-written stamp would only prove the port agrees with itself. The artifacts are vendored with a trailing `.txt` and their real names restored at load time, because `format:check`, `lint`, and `typecheck:test` all glob `tests/**/*.ts` — and a single reformatted byte would invalidate the very hash the fixture exists to pin.
-- **E2E suites (`tests/e2e/*.e2e.ts`, `make test-e2e`):** no mocks — they drive a **live pipelex-api server** (`PIPELEX_E2E_BASE_URL`, default `http://localhost:8081`) and verify the diagnostic-endpoint contract end-to-end with correct and malformed MTHDS bundles: clean content lints with zero diagnostics and formats to itself (idempotently), malformed content comes back as a 200 verdict (`diagnostics[]` on lint; `diagnostics[]` plus the content echoed back unchanged with `changed: false` on format; `is_valid: false` + `validation_errors[]` on validate), and malformed formatter options surface as a 422 `ApiResponseError`. The crate suite adds the case only a live server can settle: the `kind` / `target` vocabulary is a mirror of a Python `StrEnum` in another repo, so nothing but a real call proves the two still agree — and it cross-checks that `codegen`'s `crate_fingerprint` equals the `fingerprint` `resolve` returns for the same closure. It then runs `runCodegenCheck` over the artifacts and lock the server just emitted — verbatim first (current, with the lock header's fingerprint matching the response's), then mutated in memory once per drift category, for a `.ts` target and a `.py` one. The unit suite pins the port against vendored bytes; only this pins it against the stamp and lock bytes the server writes **today**, which is the pairing that matters when a consumer commits the server's artifacts and gates CI on the SDK's verdict about them. The distinct `.e2e.ts` suffix + `vitest.e2e.config.ts` keep them out of the default `vitest run`, so `make test` / CI never require a server. An unreachable server is an error, never a silent skip, and it is reported once: `make test-e2e` preflights the target with a single `curl` and bails with a one-line message, instead of letting all three suites fail in `beforeAll` and print a socket-error dump each. The per-suite `beforeAll` guard stays as the backstop for a direct `npm run test:e2e` run. Both probe **`/v1/version`**, deliberately: a bare runner serves the origin-level `/health` and a hosted origin does not (it answers `/v1/health` instead), so `/v1/version` is the one unauthenticated route both answer — probing anything else reports a perfectly reachable hosted origin as down. The target itself is resolved once, in the Makefile, and exported along with `PIPELEX_API_KEY`, so the URL the preflight checks is always the URL the suites call: the shell environment (or a `make test-e2e PIPELEX_E2E_BASE_URL=...` override) wins, then `.env`, then `http://localhost:8081`. The run echoes the resolved target before it starts, since `.env` can now supply it invisibly. Note that a hosted origin cannot run the whole suite today — `lint` and `format` are not exposed there (gateway `403`), while the protocol, build and crate routes all answer — so a fully green run means a local runner. The upload-grant suite needs the same hosted platform: it requests a grant, sends a small PDF with `uploadWithGrant`, reads the bytes back equal through `fetchArtifact`, and proves the refusals the error mapping reads — the `412` of a used grant, the `403 SignatureDoesNotMatch` of a file one byte longer than declared, and the `413` of a declared size over the cap. It runs in Node, whose `fetch` sends no CORS preflight, so the bucket's CORS rule is proven by a page, not by it. The artifact suite is the mirror image: it uploads a file with `prepareInputs`, runs a pass-through and brings the file back down with `downloadArtifacts` over `working_memory`, bytes equal, so it needs a hosted platform carrying the upload, run-lifecycle and bulk resolve routes and a key for it, and it fails honestly rather than skipping against a bare runner.
+- **E2E suites (`tests/e2e/*.e2e.ts`, `make test-e2e`):** no mocks — they drive a **live pipelex-api server** (`PIPELEX_E2E_BASE_URL`, default `http://localhost:8081`) and verify the diagnostic-endpoint contract end-to-end with correct and malformed MTHDS bundles: clean content lints with zero diagnostics and formats to itself (idempotently), malformed content comes back as a 200 verdict (`diagnostics[]` on lint; `diagnostics[]` plus the content echoed back unchanged with `changed: false` on format; `is_valid: false` + `validation_errors[]` on validate), and malformed formatter options surface as a 422 `ApiResponseError`. The crate suite adds the case only a live server can settle: the `kind` / `target` vocabulary is a mirror of a Python `StrEnum` in another repo, so nothing but a real call proves the two still agree — and it cross-checks that `codegen`'s `crate_fingerprint` equals the `fingerprint` `resolve` returns for the same closure. It then runs `runCodegenCheck` over the artifacts and lock the server just emitted — verbatim first (current, with the lock header's fingerprint matching the response's), then mutated in memory once per drift category, for a `.ts` target and a `.py` one. The unit suite pins the port against vendored bytes; only this pins it against the stamp and lock bytes the server writes **today**, which is the pairing that matters when a consumer commits the server's artifacts and gates CI on the SDK's verdict about them. The same suite drives `pipeIo` across all three selectors: the artifacts under the resolved ref, their equality with `validate`'s views for the same closure, the `include_files` echo, `all_pipes` on a method with no entry pipe, a `method_ref` package's manifest entry pipe, the invalid arm, and the typed selection `422`. Its `method_id` case, like the `prepareInputs` suite's, runs only when `PIPELEX_E2E_METHOD_ID` names a stored method the key's organization owns, since only a hosted origin can resolve one, and is skipped otherwise. The distinct `.e2e.ts` suffix + `vitest.e2e.config.ts` keep them out of the default `vitest run`, so `make test` / CI never require a server. An unreachable server is an error, never a silent skip, and it is reported once: `make test-e2e` preflights the target with a single `curl` and bails with a one-line message, instead of letting all three suites fail in `beforeAll` and print a socket-error dump each. The per-suite `beforeAll` guard stays as the backstop for a direct `npm run test:e2e` run. Both probe **`/v1/version`**, deliberately: a bare runner serves the origin-level `/health` and a hosted origin does not (it answers `/v1/health` instead), so `/v1/version` is the one unauthenticated route both answer — probing anything else reports a perfectly reachable hosted origin as down. The target itself is resolved once, in the Makefile, and exported along with `PIPELEX_API_KEY`, so the URL the preflight checks is always the URL the suites call: the shell environment (or a `make test-e2e PIPELEX_E2E_BASE_URL=...` override) wins, then `.env`, then `http://localhost:8081`. The run echoes the resolved target before it starts, since `.env` can now supply it invisibly. Note that a hosted origin cannot run the whole suite today — `lint` and `format` are not exposed there (gateway `403`), while the protocol and build routes and the `resolve` and `codegen` crate routes answer, and `pipe-io` answers only once the platform's proxy and the gateway list it — so a fully green run means a local runner. The upload-grant suite needs the same hosted platform: it requests a grant, sends a small PDF with `uploadWithGrant`, reads the bytes back equal through `fetchArtifact`, and proves the refusals the error mapping reads — the `412` of a used grant, the `403 SignatureDoesNotMatch` of a file one byte longer than declared, and the `413` of a declared size over the cap. It runs in Node, whose `fetch` sends no CORS preflight, so the bucket's CORS rule is proven by a page, not by it. The artifact suite is the mirror image: it uploads a file with `prepareInputs`, runs a pass-through and brings the file back down with `downloadArtifacts` over `working_memory`, bytes equal, so it needs a hosted platform carrying `POST /v1/pipe-io` (which `prepareInputs` reads), the upload, run-lifecycle and bulk resolve routes, and a key for it, and it fails honestly rather than skipping against a bare runner.

package/docs/artifact-download.md

@@ -104,7 +104,7 @@ if (!verdict.all_saved) {
 
 **Node-only**, like the path-string arm of `uploadFile`: it writes to a filesystem, and says so with an `ArtifactOperationError` anywhere else.
 
-**Where it reads from.** Exactly one of `run_id` and `results`. A `run_id` re-reads the results through `getRunResult`, so a completed run is downloadable days later from its id alone; a `RunResults` already in hand is read as it is, with no request. `scope` picks the artifact walked for references: `main_stuff` (the default) is the run's output, and `working_memory` is the opt-in that also brings down the echoed inputs and every intermediate stuff — it is read off the parsed results body, whether or not `RunResults` declares the field.
+**Where it reads from.** Exactly one of `run_id` and `results`. A `run_id` re-reads the results through `getRunResult`, asking only for the scope's artifact, so a completed run is downloadable days later from its id alone; a `RunResults` already in hand is read as it is, with no request. `scope` picks the artifact walked for references: `main_stuff` (the default) is the run's output, and `working_memory` is the opt-in that also brings down the echoed inputs and every intermediate stuff — it is read off the parsed results body, whether or not `RunResults` declares the field.
 
 **How it downloads.** The whole set is resolved through the bulk route ahead of the workers, then a bounded number of workers (`concurrency`, default 4) each take the next reference and fetch it → open its file with `wx` → stream the body in. Resolution is just-in-time where it matters: a link that has expired by the time its worker reaches it — a large set downloaded a few at a time can outlive the fifteen-minute link — is resolved again for that reference alone, so no fetch ever runs on a stale signature.
 

package/docs/build-routes.md

@@ -2,7 +2,7 @@
 
 The three build projections — `buildInputs`, `buildOutput`, `buildRunner` — take a **closure** of MTHDS files, select **one pipe** in it, and project something from that pipe. They are Pipelex API extensions, not MTHDS Protocol routes: the standard specifies no projection surface, so these are ours to shape.
 
-> **`prepareInputs` no longer reads the inputs template.** It resolves its signature from the input-form descriptor on the validate report (see [input-preparation.md](./input-preparation.md#signature-driven-asset-identification)), so nothing inside this SDK calls these routes. They are kept for external consumers, and their retirement across the workspace is tracked separately.
+> **`prepareInputs` no longer reads the inputs template.** It reads its signature from the input-form descriptor `POST /v1/pipe-io` returns (see [input-preparation.md](./input-preparation.md#signature-driven-asset-identification)), so nothing inside this SDK calls these routes. They are kept for external consumers, and their retirement across the workspace is tracked separately.
 
 ## The shared envelope
 

package/docs/crate-routes.md

@@ -1,16 +1,18 @@
-# Crate routes (`/v1/resolve`, `/v1/codegen`)
+# Crate routes (`/v1/resolve`, `/v1/codegen`, `/v1/pipe-io`)
 
-Two routes project a **closure** of MTHDS files into the artifacts downstream tooling actually consumes: `resolve` emits the **normalized library crate**, and `codegen` projects that crate into **stamped typed artifacts** plus their lock. Like the [build routes](./build-routes.md), they are Pipelex API extensions rather than MTHDS Protocol operations — but note the ownership split: the _crate_ is standard-owned (the MTHDS Library Crate Format), while the HTTP surface serving it, and every type projection on top of it, are ours.
+Three routes project a **closure** of MTHDS files into the artifacts downstream tooling actually consumes: `resolve` emits the **normalized library crate**, `codegen` projects that crate into **stamped typed artifacts** plus their lock, and `pipeIo` derives a method's **pipe I/O contracts, input form and output form** without a dry run. Like the [build routes](./build-routes.md), they are Pipelex API extensions rather than MTHDS Protocol operations — but note the ownership split: the _crate_ and the three I/O artifacts are standard-owned (the MTHDS Library Crate Format, and the standard's pipe I/O contracts and form descriptors), while the HTTP surface serving them, and every type projection on top of the crate, are ours.
 
-> **Both crate routes are served on every hosted origin, and by any `pipelex-api` runner.** On the hosted plane a route is reachable only when the gateway's API-key allowlist and the platform's tooling proxy both list its path — each enumerates routes explicitly, and an unlisted path answers a gateway `403 {"message":"Forbidden"}`, refused before any service sees the request, so not even an RFC 7807 problem body. `resolve` and `codegen` are listed by both.
+> **`resolve` and `codegen` are served on every hosted origin, and by any `pipelex-api` runner.** On the hosted plane a route is reachable only when the gateway's API-key allowlist and the platform's tooling proxy both list its path — each enumerates routes explicitly, and an unlisted path answers a gateway `403 {"message":"Forbidden"}`, refused before any service sees the request, so not even an RFC 7807 problem body. `resolve` and `codegen` are listed by both.
 >
 > Measured 2026-08-23 against `api.pipelex.com` (`pipelex-hosted@0.10.1`) with a real API key: an empty body to either route comes back as an RFC 7807 `422` naming the missing fields — a request-shape verdict only a route that reached the service can produce — and a real closure comes back as a `200` carrying the artifacts. `api-dev.pipelex.com` has served them since 2026-08-13. Use a key when re-measuring: unauthenticated, every path answers `401` whether or not it is allowlisted, so a keyless probe cannot tell the two states apart.
 >
-> **`lint` / `format` are the exception and answer `403` on both origins.** That blocks nothing, because linting and formatting `.mthds` are toolchain capabilities rather than hosted ones: `plxt` carries both, and the post-edit hook this repo builds (`npm run build:hook`, vendored into `pipelex-plugins`) runs them offline through `@pipelex/tools-wasm` with no credentials — from the published package, `client.lint` / `client.format` are the documented fallback, against a runner. Exposing them on the hosted origins is a known, non-critical item on the platform's list, tracked in [`wip/hosted-exposure-crate-and-tools-routes.md`](https://github.com/Pipelex/pipelex-sdk-js/blob/main/wip/hosted-exposure-crate-and-tools-routes.md).
+> **`pipe-io` is newer.** A runner serves it from `pipelex-api` v0.33.0, and types its pipe-selection refusals `EntryPipeNotFoundError` and `EntryPipeAmbiguousError` from v0.33.1, which is what `prepareInputs` branches on (see [What throws](#what-throws)); v0.33.0 typed them as the generic `ValidationError`. A hosted origin serves the route once the platform's tooling proxy and the gateway's allowlist both list it, and until then it answers the gateway's `403 {"message":"Forbidden"}`. `prepareInputs` reads this route, so it needs an origin that serves it.
+>
+> **`lint` / `format` are the exception and answer `403` on both origins.** That blocks nothing, because linting and formatting `.mthds` are toolchain capabilities rather than hosted ones: `plxt` carries both, and the post-edit hook this repo builds (`npm run build:hook`, vendored into `pipelex-plugins`) runs them offline through `@pipelex/tools-wasm` with no credentials — from the published package, `client.lint` / `client.format` are the documented fallback, against a runner. Exposing them on the hosted origins is a known, non-critical item on the platform's list, tracked in the workspace ledger as L-260929-b58f26.
 
 ## The shared envelope
 
-Both take the same closure selector — inline `files`, a `method_ref`, or a hosted `method_id`, exactly one (the strict tooling XOR):
+All three take the same closure selector — inline `files`, a `method_ref`, or a hosted `method_id`, exactly one (the strict tooling XOR):
 
 ```ts
 interface CrateRequestBase {
@@ -31,7 +33,9 @@ An **address-form** `method_ref` (`github.com/<owner>/<repo>[/<selector>][@<tag>
 
 Supplying **no** selector or **more than one** is a request-shape `422` — the tooling routes are stateless, so there is no linkage exception; a second selector could only be ignored, which is the worst contract of the three. The SDK does not model the XOR in the type system — the union would force the overwhelmingly common `{ files }` call site to pick a branch for no gain, and the server's answer is a typed `ApiResponseError` either way.
 
-The old advice to expand a stored method client-side (`resolve({ files: await client.getMethodClosure(methodId) })`) remains valid — `getMethodClosure` stays public as the local expansion utility, and it is what a caller uses against a bare runner or on the routes with no by-id form (`/v1/build/*`, `prepareInputs`).
+The old advice to expand a stored method client-side (`resolve({ files: await client.getMethodClosure(methodId) })`) remains valid — `getMethodClosure` stays public as the local expansion utility, and it is what a caller uses against a bare runner or on the routes with no by-id form (`/v1/build/*`).
+
+A `method_ref` makes the server fetch the repository before it answers, so the client gives a call carrying one a fetch-sized budget (three minutes) instead of the 30-second default. On the hosted API the gateway caps every request at about 30 seconds whatever the client allows, so a cold `method_ref` clone can answer a `502`; retrying clears it once the runner has cached the clone.
 
 ## `resolve` — the normalized crate
 
@@ -83,23 +87,56 @@ The SDK stays transport-only: it hands you the artifacts and does not write file
 
 `kind: "types"` is concept-set-wide, so passing `pipe_ref` alongside it is a request-shape **`422`**. Silently ignoring the selector would mislead a caller into believing the artifacts were narrowed to one pipe. The field exists on the request for the future per-pipe kinds.
 
+## `pipeIo` — a method's I/O artifacts, without a validation
+
+`pipeIo` resolves the closure the way `resolve` does, selects a pipe, and returns the method's three I/O artifacts — the standard's `PipeIOContracts`, `InputForm` and `OutputForm`, typed by importing them from `mthds/protocol` — with the selection and the runnability facts beside them. It runs **no dry-run sweep**, so it costs one load and one derivation where `validate` dry-runs every pipe. Read it to show a method, to prepare its inputs (`prepareInputs` does) or to generate types for it; stay on `validate` for the dry-run verdict or the dry-run graph.
+
+```ts
+const result = await client.pipeIo({ method_ref: "github.com/Pipelex/methods/documents" });
+if (!result.is_valid) return;
+
+const form = result.input_form[result.pipe_ref!]; // the entry pipe's input form
+```
+
+The request is the shared envelope plus three optional fields, posted verbatim:
+
+| Field           | Default | Meaning                                                                                                   |
+| --------------- | ------- | --------------------------------------------------------------------------------------------------------- |
+| `pipe_ref`      | none    | The qualified `domain.pipe_code` to describe. Omitted, the server's selection chain decides.              |
+| `all_pipes`     | `false` | Describe every pipe the closure loads instead of the selected one.                                        |
+| `include_files` | `false` | Echo the resolved closure's `.mthds` files as `files`, in the request's own `{ content, source? }` shape. |
+
+**Selection.** The request's `pipe_ref`; else a fetched package's manifest `main_pipe`; else the closure's `main_pipe`, when exactly one domain declares one. A `pipe_ref` that names no pipe, and — without `all_pipes` — a chain that finds no entry pipe or several, are refused with a `422` whose `errorType` names the refusal (see [What throws](#what-throws)). The server does not refuse a bare code yet: a bare ref that matches one pipe resolves across domains, and the answer reports the qualified ref. With `all_pipes: true` the route never refuses for want of an entry pipe, so a method that declares none is still describable.
+
+The valid arm (`PipeIOValidReport`):
+
+- `pipe_ref` — the qualified ref the selection resolved, read off the resolved pipe and never echoed from the request; `null` only under `all_pipes` when nothing resolves.
+- `pipe_io_contracts`, `input_form`, `output_form` — the three maps, sharing one key set: the resolved `pipe_ref` alone by default, every pipe under `all_pipes`. For a closure `validate` also accepts, each equals `validate`'s same-named field restricted to the same keys, null members such as a contract's `item_count: null` included.
+- `default_pipe_ref` — the method's own entry pipe, the selection chain without the request's `pipe_ref`, or `null` when that chain finds none or several. A request that omits `pipe_ref` always answers `pipe_ref === default_pipe_ref`. It is **not** `validate`'s field of the same name, which is the run default and names the first of several declaring domains where this route refuses to choose.
+- `pending_signatures`, `is_runnable` — what they mean on `validate`, read off the loaded library with no dry run, so a method whose dry run would fail is still reported runnable here.
+- `files` — present only with `include_files: true`: the request's files for inline `files`, the fetched package's `.mthds` files under their package-relative paths for a `method_ref`, the stored files under their stored names for a hosted `method_id`. Absent, not empty, otherwise.
+
+`is_valid: true` means what it means on `resolve`: the closure parsed, loaded and passed static validation. The invalid arm is the shared `CrateInvalidReport`, and it carries no artifacts, no selection and no files, whatever the request asked for.
+
 ## The response is a verdict, not a payload
 
-Both routes return a **discriminated 200**, the same discipline as `validate` and the build routes: an unresolvable closure is the _successful product_ of the call (the request was well-formed; the library was not), so it rides a 200 with `is_valid: false` and the shared `CrateInvalidReport` — the very same invalid arm the build routes return, carrying the same structured `validation_errors[]`.
+All three routes return a **discriminated 200**, the same discipline as `validate` and the build routes: an unresolvable closure is the _successful product_ of the call (the request was well-formed; the library was not), so it rides a 200 with `is_valid: false` and the shared `CrateInvalidReport` — the very same invalid arm the build routes return, carrying the same structured `validation_errors[]`.
 
 **Branch on `is_valid` before reading the arm.** A consumer that only catches throws will render a success over an unusable result, because nothing threw.
 
 ## What throws
 
-Only a **no-verdict** condition, as the typed `ApiResponseError` — branch on its `status`, never on its message:
+Only a **no-verdict** condition, as the typed `ApiResponseError` — branch on its `status` and `errorType`, never on its message:
 
 | Status        | Cause                                                                                                     |
 | ------------- | --------------------------------------------------------------------------------------------------------- |
-| `422`         | Request shape: no closure selector or more than one, an unknown `kind`/`target`, a `pipe_ref` on `kind: "types"`, a stored method with no MTHDS source. |
-| `404`         | Unknown or foreign-org `method_id` (indistinguishable by design).                                           |
+| `422`         | Request shape: no closure selector or more than one, an over-limit file, an unknown `kind`/`target`, a `pipe_ref` on `kind: "types"`, a fetched package with no `.mthds` file, a stored method with no MTHDS source. |
+| `422`         | On `pipeIo`, a pipe selection the route refuses: `errorType` `EntryPipeNotFoundError` for a `pipe_ref` that names no pipe or a method with no entry pipe, `EntryPipeAmbiguousError` for a code that matches pipes in several domains or several `main_pipe` declarations. The `serverMessage` names the candidates where there are some. |
+| `404`         | Unknown or foreign-org `method_id` (indistinguishable by design); no package at a `method_ref` address.     |
 | `501`         | Registry-form `method_ref` — reserved, not implemented (the address form resolves).                         |
-| `401` / `403` | Auth.                                                                                                       |
-| `5xx`         | Server fault.                                                                                               |
+| `401` / `403` | Auth, or on a hosted origin a route the gateway does not list yet (`403 {"message":"Forbidden"}`).          |
+| `502`         | On a hosted origin, a cold `method_ref` clone that outran the gateway's 30-second cap; a retry clears it.   |
+| `5xx`         | Server fault, including on `pipeIo` a pipe whose artifacts cannot be derived.                               |
 
 Note the split, same as the build routes: a bad **closure** is a 200 verdict; a bad **request** throws.
 

package/docs/errors.md

@@ -15,7 +15,8 @@ The report reaches you in four places, always the same object:
 | `getRunResult(runId)` | the `failed` arm: `{ state: "failed", pipeline_run_id, status, message, error }` |
 | `waitForResult`, `startAndWaitForResult` on the hosted API, `downloadArtifacts({ run_id })` | a thrown `RunFailedError` with `runId`, `status`, `message` and `error` |
 | `getRunStatus(runId)` | `RunRead.error` on the run record |
-| `listRuns`, `iterateRuns`, `getRunDetail` | `PipelineRun.error` on each run record |
+| `listRuns`, `iterateRuns` | `RunHistoryItem.error` on each history row |
+| `getRunDetail` | `PipelineRun.error` on the run record |
 
 **A bare runner has no durable run, so its failure is a refused request.** Against a bare `pipelex-api` runner, `startAndWaitForResult` runs the method with the blocking `execute`, and a run that fails there answers a non-2xx problem document: the call throws an `ApiResponseError`, not a `RunFailedError`, and the same classification rides its members (`errorDomain`, `type`, `retryable`, `userAction`, `errorType`, `model`, …) as described [below](#a-refused-request--apiresponseerror). A caller that must work against both catches both, as the example does.
 

package/docs/input-preparation.md

@@ -2,7 +2,7 @@
 
 > **Status: implemented** (`src/upload.ts`, `src/upload-grant.ts`, `src/prepare-inputs.ts`). This document records the contract, and is the public account of it. The raw `upload()` primitive described in [architecture.md](./architecture.md) is the wire call `uploadFile` and `prepareInputs` build on.
 >
-> **Current scope.** `prepareInputs` names the method the same three ways every other method-taking operation does — inline `files`, a `method_ref` address, or a stored `method_id` — exactly one per call, all three resolved server-side by the one `validate` call it composes. One piece remains deliberately deferred and additive (it does not change this contract): the opt-in ingest of `http(s)` URLs into storage — for now an `http(s)` URL at a file position always passes through unchanged.
+> **Current scope.** `prepareInputs` names the method the same three ways every other method-taking operation does — inline `files`, a `method_ref` address, or a stored `method_id` — exactly one per call, all three resolved server-side by the one `POST /v1/pipe-io` call it composes. One piece remains deliberately deferred and additive (it does not change this contract): the opt-in ingest of `http(s)` URLs into storage — for now an `http(s)` URL at a file position always passes through unchanged.
 
 ## Why this exists
 
@@ -117,30 +117,27 @@ Nothing is expanded client-side: `method_id` here is a **pass-through**, the sam
 
 #### Where the signature comes from
 
-One `POST /v1/validate` per call, whatever the selector, asking for the **input-form descriptor**:
+One `POST /v1/pipe-io` per call, whatever the selector, through the client's own `pipeIo` (see [crate-routes.md](./crate-routes.md#pipeio--a-methods-io-artifacts-without-a-validation)):
 
 ```
-validate(<selector or contents>, allowSignatures = true, …, views = ["input_form"])
+pipeIo({ <files | method_ref | method_id>, pipe_ref? })
 ```
 
-`allowSignatures: true` is deliberate. Preparation needs a pipe's *declared* inputs, and a bundle mid-authoring with an unresolved signature somewhere else must not be refused inputs for a pipe whose inputs are declared — whether the bundle runs is the run's verdict, not preparation's. An `is_valid: false` verdict still means the closure does not load, which is a preparation failure.
+The route resolves the closure, **selects the pipe**, and answers with that pipe's **input-form descriptor** under the qualified ref it resolved. It runs no dry run, so preparing inputs costs one static load rather than a validation of the whole method. It also refuses nothing for a pending signature elsewhere in the method: preparation needs a pipe's *declared* inputs, and whether the method runs is the run's verdict, not preparation's. An `is_valid: false` verdict still means the closure does not load, which is a preparation failure.
 
-A `method_ref` makes the server clone a repository first; `validate` needs no special budget for it, because the route already defaults to the 20-minute execute ceiling.
+A `method_ref` makes the server clone a repository first, so the call gets the crate routes' fetch-sized budget. On the hosted API the gateway caps a request at about 30 seconds whatever the client allows, so a cold clone can answer a `502` `ApiResponseError`; a retry clears it once the runner has cached the clone.
+
+`prepareInputs` needs an API that serves `POST /v1/pipe-io`. Against one that does not, the route's `404` (a runner) or `403` (a hosted origin whose gateway does not list it yet) propagates as an `ApiResponseError`. It also needs the route's pipe-selection refusals typed, as step 3 of "Pipe selection" below describes, which `pipelex-api` does from v0.33.1: v0.33.0, the release that added the route, typed them as the generic `ValidationError`, so against it a refused selection reaches the caller as that `ApiResponseError` rather than an `InputPreparationError`.
 
 #### Pipe selection
 
-`validate` has no pipe selector — its report describes every pipe, keyed by qualified `pipe_ref` — so the helper picks one, in this order:
+The route selects the pipe, and the ref it answers with is the one prepared — the helper derives no default of its own:
 
-1. **`pipe_ref` when given.** Qualified-only: `domain.pipe_code`. A bare code, or a ref the method does not declare, is an `InputPreparationError` listing the qualified refs — one step to fix. The helper never grows a searched `pipe_code`: search is a run-route affordance, and the descriptor is keyed by qualified refs. An `alias->domain.pipe_code` ref is refused for a different reason: the descriptor describes the method's *own* pipes, and an alias names one belonging to a dependency package, which the report never carries a descriptor for — the run route takes such a ref, preparation cannot.
-2. **The report's resolved default** (`default_pipe_ref`): the qualified ref of the pipe a selector-less run of *this same method* would execute, manifest-aware for a fetched package. It is read on the field's **presence**, never on its truthiness, and has three arms:
-   - a **non-empty string** is the default and is walked — and if the descriptor does not describe it, preparation refuses rather than falling through, since one report keyed by one pipe set naming a pipe it does not describe is the report contradicting itself;
-   - a **stated `null`** — or any stated value that is not a non-empty string — is the server saying it determined no entry pipe: no blueprint declares a `main_pipe`, or the package manifest names a pipe the closure does not declare or declares in several domains. A run naming no pipe is refused in exactly those cases, so preparation refuses too and asks for `pipe_ref`. Steps 3 and 4 are **not** consulted behind it;
-   - an **absent field** means the runner predates it, and only then do steps 3 and 4 stand.
-3. **The bundle's declared `main_pipe`** — behind an absent field only — read defensively from the opaque `bundle_blueprint` and qualified by its `domain`.
-4. **The single pipe** — behind an absent field only — when the method declares exactly one.
-5. Otherwise an `InputPreparationError` naming the candidates and asking for `pipe_ref`.
+1. **`pipe_ref` when given.** Qualified-only: `domain.pipe_code`, normalized (trimmed, and blank is absent) and checked **before any request**. A bare code is refused with an `InputPreparationError` asking for the qualified form, since a request names a pipe by its qualified ref and search is a run-route affordance this helper never grows. An `alias->domain.pipe_code` ref is refused for a different reason: the alias names a dependency package's pipe, and preparation covers the method's *own* pipes — the crate routes do not load an address-based dependency at all. The run route takes such a ref; preparation cannot.
+2. **Otherwise the method's own entry pipe**, chosen by the route: a fetched package's manifest `main_pipe`, else the closure's `main_pipe` when exactly one domain declares one. A published package that names its entry pipe in `METHODS.toml` alone — `github.com/Pipelex/methods/documents` does — therefore prepares with no `pipe_ref`.
+3. **A selection the route refuses** — a `pipe_ref` that names no pipe of the method, a method with no entry pipe, one whose domains declare several, a manifest `main_pipe` the closure does not declare or declares in several domains — comes back as a `422` whose `errorType` is `EntryPipeNotFoundError` or `EntryPipeAmbiguousError`. The helper turns exactly those into an `InputPreparationError` carrying the server's detail, which names the candidates where there are some, with the `ApiResponseError` as its `cause`. The run route refuses a selector-less run in the same cases, so preparation never succeeds for a pipe the run would not take.
 
-> **The manifest-only `main_pipe` gap is what step 2 closes.** A published package may name its entry pipe in `METHODS.toml` alone — `github.com/Pipelex/methods/documents` and `.../image_generation` do — and the validate report carries no manifest; the runner resolves the manifest server-side and states the result in `default_pipe_ref`, so such a package prepares with no `pipe_ref`. A manifest naming a pipe the closure does not declare, or declares in several domains, is the stated `null` instead: the run would fail on that code, and preparation says so before anything is uploaded.
+There is no single-pipe fallback: a method that declares one pipe and no `main_pipe` is refused a selector-less run, and a selector-less preparation with it.
 
 ## `getMethodClosure` — the explicit expansion utility
 
@@ -156,7 +153,7 @@ Reach for it whenever you want the files **in hand** — to edit, to diff, or to
 client.buildInputs({ files: await client.getMethodClosure("mt_…"), pipe_ref?, format?, explicit? })
 ```
 
-`/v1/build/*` is the last such family, and it is being retired. Every other method-taking operation — `execute` / `start`, `validate` / `resolve` / `codegen`, and now `prepareInputs` — takes a `method_id` **natively**, as a server pass-through: the platform resolves it and injects the stored source before the runner sees the request (see [architecture.md → hosted run extensions](./architecture.md#hosted-run-extensions-method_id)). That is the uniform rule — *an id option is always a server pass-through, and any client-side expansion is the caller's own explicit call*. An untyped caller still passing `method_id` to `buildInputs` gets a teaching error naming this migration (`PipelineRequestError`); for typed callers the field is a compile error.
+`/v1/build/*` is the last such family, and it is being retired. Every other method-taking operation — `execute` / `start`, `validate` / `resolve` / `codegen` / `pipeIo`, and `prepareInputs` — takes a `method_id` **natively**, as a server pass-through: the platform resolves it and injects the stored source before the runner sees the request (see [architecture.md → hosted run extensions](./architecture.md#hosted-run-extensions-method_id)). That is the uniform rule — *an id option is always a server pass-through, and any client-side expansion is the caller's own explicit call*. An untyped caller still passing `method_id` to `buildInputs` gets a teaching error naming this migration (`PipelineRequestError`); for typed callers the field is a compile error.
 
 - **Requires an API key.** The methods catalog is org-scoped to the key's org, so resolution only works with an authenticated Pipelex-product key.
 - **Unknown or foreign-org id → the `getMethod` `404`** (`ApiResponseError`, `code: "not_found"`), propagated unchanged. An id from another org is indistinguishable from a nonexistent one — both 404.
@@ -168,7 +165,7 @@ client.buildInputs({ files: await client.getMethodClosure("mt_…"), pipe_ref?,
 methodSourceToContents(mthds: string) → string[]
 ```
 
-A stored method's `mthds` field is **polymorphic**: it is either the raw single-bundle source, or a JSON array of `{ name, content }` file entries (a multi-file closure). `methodSourceToContents` is the one canonical parser that turns either shape into a flat list of file contents — a verbatim port of the platform's `_method_source_to_contents`, so the SDK and the runtime read a stored source identically. It is exported from the barrel for consumers (e.g. `pipelex-mcp`) that need the parse without the fetch. Blank entries are dropped; a source that parses to no non-blank content yields `[]`, which is what `getMethodClosure` turns into `EmptyMethodSourceError`.
+A stored method's `mthds` field is **polymorphic**: it is either the raw single-bundle source, or a JSON array of `{ name, content }` file entries (a multi-file closure). `methodSourceToContents` is the one canonical parser that turns either shape into a flat list of file contents — a verbatim port of the platform's `method_source_to_contents`, so the SDK and the runtime read a stored source identically. It is exported from the barrel for consumers (e.g. `pipelex-mcp`) that need the parse without the fetch. Blank entries are dropped; a source that parses to no non-blank content yields `[]`, which is what `getMethodClosure` turns into `EmptyMethodSourceError`.
 
 ## Compact or explicit-envelope inputs
 
@@ -183,7 +180,7 @@ When a value is an envelope (a plain object whose keys are **exactly** `concept`
 
 The SDK **must not** guess that every string resembling a path is an asset — that would make ordinary text inputs environment-dependent and could upload unintended files. Interpretation comes from the method's **declared signature**, never from a value's shape alone. This mirrors the runtime's own top-down interpretation (`InputShaper`) combined with the file-reference resolution of `input_normalizer` in `pipelex`, so local and hosted execution read the same inputs the same way. (A caller value may be compact or the explicit `{ concept, content }` envelope — see "[Compact or explicit-envelope inputs](#compact-or-explicit-envelope-inputs)"; the walk below is over the compact value, i.e. the envelope's `content`.)
 
-The declared signature is the **input-form descriptor** — the MTHDS standard's `input_form` artifact, carried on the validate report under `views: ["input_form"]`. It states the kind of every input at every depth, so a file position is a *fact of the method*, never an inference from the value in front of it.
+The declared signature is the **input-form descriptor** — the MTHDS standard's `input_form` artifact, which `POST /v1/pipe-io` answers with for the pipe it selected. It states the kind of every input at every depth, so a file position is a *fact of the method*, never an inference from the value in front of it.
 
 The walk is discriminated on the node's `kind`:
 
@@ -223,7 +220,7 @@ Upload is a **hosted Pipelex-product capability**, even though the SDK can be po
 
 The contract distinguishes these semantic outcomes, each a typed subclass of `InputPreparationError` (catch the base to handle any preparation failure, a subclass to branch on category):
 
-- **an unresolvable signature** (`InputPreparationError`) — the closure did not validate (`is_valid: false`, carrying the first error's message), the report carried no `input_form` descriptor, or no pipe could be chosen (an unknown or bare `pipe_ref`, or no default). A *no-verdict* condition from `/v1/validate` — a malformed selector, an unknown or foreign-org `method_id`, no package at the address, auth, a server fault — stays an `ApiResponseError` and propagates unchanged;
+- **an unresolvable signature** (`InputPreparationError`) — a bare or alias-qualified `pipe_ref`, refused before any request; a closure that does not load (`is_valid: false`, carrying the first error's message); or a pipe selection the route refused (its `422` typed `EntryPipeNotFoundError` or `EntryPipeAmbiguousError`, carrying the server's detail). Any other failure from `/v1/pipe-io` — a malformed selector, an over-limit file, an unknown or foreign-org `method_id`, no package at the address or one with no `.mthds` file, auth, a server fault, a deployment that does not serve the route — stays an `ApiResponseError` and propagates unchanged;
 - **empty method source** (`EmptyMethodSourceError`, carries `methodId`) — `getMethodClosure` found the stored method but its `mthds` source parses to nothing (the row exists, no runnable source yet). Distinct from the `getMethod` `404` for an unknown/foreign id, which stays an `ApiResponseError`. `prepareInputs` never raises it: it hands the id to the server, which answers a sourceless method with a `422`;
 - **invalid local source** (`InvalidLocalSourceError`) — missing, unreadable, or a path string outside Node;
 - **rejected asset** (`RejectedAssetError`) — the server refused it (e.g. a `413` past the service-defined size cap — see "Storage policy" — surfaced as a clear rejection with `code` `too_large`, not a raw transport error), or storage refused an upload with a grant (a used or expired grant, or a file that differs from what the grant signed — see the table above for each `code`);

package/docs/run-results.md

@@ -17,7 +17,7 @@ console.log(results.main_stuff, summarizeUsage(results).total_cost_usd);
 | field | type | hosted (durable) path | bare-runner (blocking) path |
 |---|---|---|---|
 | `pipeline_run_id` | `string` | the run store's id | the runner's own id for the call |
-| `main_stuff` | `unknown` | the `main_stuff.json` artifact | resolved out of the returned working memory |
+| `main_stuff` | `unknown` (optional) | the `main_stuff.json` artifact | resolved out of the returned working memory |
 | `working_memory` | `DictWorkingMemory \| null` | the `working_memory.json` artifact | lifted off `pipe_output` |
 | `graph_spec` | `unknown` | the `graphspec.json` artifact | lifted off `pipe_output` |
 | `graph_assembly_error` | `string \| null \| undefined` | absent until the platform relays it | lifted off `pipe_output` |
@@ -29,6 +29,19 @@ console.log(results.main_stuff, summarizeUsage(results).total_cost_usd);
 | `usage_assembly_error` | `string \| null` | relayed | lifted off `pipe_output` |
 | `pipe_output` | `DictPipeOutput \| null \| undefined` | absent | the runner's whole native output |
 
+## Reading only some artifacts — `artifacts`
+
+A results read on the hosted path fetches every artifact from storage and re-signs every link inside them, which is wasted work for a caller that wants one of them, such as a history view showing a run's output or a usage panel. `getRunResult(runId, { artifacts })` narrows the read: the SDK sends the names as one comma-separated `?artifacts=` parameter and the platform reads, re-signs and returns only those. The names are `RUN_RESULT_ARTIFACTS` — `graph_spec`, `pipe_io_contracts`, `input_form`, `output_form`, `main_stuff`, `working_memory` and `tokens_usages` — typed as `RunResultArtifact`, and `tokens_usages` brings `usage_assembly_error` with it. Without the option every artifact is read.
+
+```ts
+const state = await client.getRunResult(runId, { artifacts: ["main_stuff", "tokens_usages"] });
+const results = await client.waitForResult(runId, { artifacts: ["working_memory"] });
+```
+
+**Absent versus null.** With a selection, an artifact the caller did not name is absent from the result, reading `undefined`, while one it named that the run never wrote is `null`. So `undefined` means "not asked for" and `null` means "asked for, not there". That is also why `main_stuff` is declared optional: it is never null on a completed run, but it is absent from a read that left it out, and `MissingMainStuffError` is thrown only by a read that asked for it — one with no selection, or one naming `main_stuff`.
+
+An empty selection or a name outside the list is a `RangeError` thrown before any request, the same refusal the platform would answer with a `400`. `waitForResult` and `startAndWaitForResult` take the same `artifacts` in their poll options and send it on every poll; `startAndWaitForResult` refuses a bad selection before it starts the run. Against a bare runner the selection has nothing to narrow: the blocking fallback returns every artifact the runner sent. `downloadArtifacts` given a `run_id` asks for its scope's artifact alone.
+
 ## `pipeline_run_id` — the durable handle
 
 The run id is what makes a run readable after the process that started it has gone. `start` returns it in its acknowledgement before the run finishes, and every lifecycle read takes it: `getRunStatus(runId)` for the status row, `getRunResult(runId)` for a single result lookup, `waitForResult(runId)` to resume polling a run an earlier session started. It is also what a `RunTimeoutError` leaves you with — the run keeps executing server-side, so the timeout is a reason to re-poll by id, not a reason to run the method again.
@@ -45,7 +58,7 @@ Against a bare runner the id identifies the call the runner just answered, but t
 
 ## `main_stuff` — the output
 
-`main_stuff` is the resolved content of the run's main output and is always present for a completed run. On the hosted path it is the `main_stuff.json` artifact; on the blocking path the SDK resolves it out of the returned working memory through the response's `main_stuff_name`. Both deliver the same content shape, so there is no shape-guessing and no path-dependent branch to write. A completed run that cannot deliver one throws `MissingMainStuffError` rather than handing back a half-filled result.
+`main_stuff` is the resolved content of the run's main output and is always present for a completed run, unless the read's `artifacts` selection left it out. On the hosted path it is the `main_stuff.json` artifact; on the blocking path the SDK resolves it out of the returned working memory through the response's `main_stuff_name`. Both deliver the same content shape, so there is no shape-guessing and no path-dependent branch to write. A completed run that cannot deliver one throws `MissingMainStuffError` rather than handing back a half-filled result.
 
 It is typed `unknown` because the content is polymorphic: a structured output arrives as an object of the concept's fields, and a multiple output as the envelope `{ items: [...] }` that the runtime's `ListContent` serialises to. Every content type serialises to an object, natives included — a text output is `{ "text": "…" }` and a number `{ "number": 0 }` — so a guard written for a bare `""` or `0` never fires, and an empty multiple output is `{ "items": [] }` rather than `[]`. Narrow it where you read it, ideally through the types generated for the method rather than a hand-written cast.
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@pipelex/sdk",
-  "version": "0.26.0",
+  "version": "0.28.0",
   "description": "TypeScript SDK for the Pipelex hosted API — execute methods, manage runs, and call the product surface.",
   "license": "MIT",
   "repository": {