# EigenPal Docs

- [EigenPal docs](https://docs.eigenpal.com/index.md): Build, test, and run AI workflows for repetitive document and data tasks.
- [Choose your path](https://docs.eigenpal.com/guides/choose-your-path.md): Pick the fastest way into EigenPal based on what you are trying to do.
- [Why EigenPal](https://docs.eigenpal.com/concepts/why.md): Automate repetitive work with AI, with datasets and evaluators for measuring output quality.
- [How it works](https://docs.eigenpal.com/concepts/how-it-works.md): Build, evaluate, deploy, observe, and improve an EigenPal workflow.
- [Design principles](https://docs.eigenpal.com/concepts/principles.md): The engineering choices behind EigenPal workflows, evaluations, and artifacts.
- [Automations](https://docs.eigenpal.com/concepts/automations.md): The unified API model for running workflows and agents.
- [Workflows](https://docs.eigenpal.com/concepts/workflows.md): Directed graphs of typed steps that automate a repetitive task.
- [Agents](https://docs.eigenpal.com/concepts/agents.md): Git-backed agents for document tasks that are hard to express as a fixed step graph.
- [Evaluations](https://docs.eigenpal.com/concepts/evals.md): Test workflows like code. Datasets hold examples, evaluators score runs, experiments compare versions.
- [Reviews](https://docs.eigenpal.com/concepts/reviews.md): Human feedback for production runs: verdicts, statuses, corrections, and monitoring.
- [Artifacts as files](https://docs.eigenpal.com/concepts/artifacts.md): Workflows, datasets, evaluators, and agents are plain files you can export, import, diff, and version.
- [Workflow steps](https://docs.eigenpal.com/concepts/steps-overview.md): The complete catalog of step types, each with a code-derived config and output schema.
- [Parse Document](https://docs.eigenpal.com/steps/ai/parse.md): Extract text from documents (PDF, DOCX, images) using native extraction, OCR, or vision models
- [Extract Data](https://docs.eigenpal.com/steps/ai/extract.md): Extract structured data from text using AI with a JSON schema
- [Split Document](https://docs.eigenpal.com/steps/ai/split.md): Split a parsed document into named sections using an LLM. Consumes ai.parse output; emits per-section page ranges and text ready for downstream ai.extract via control.parallel_map.
- [Separate Documents](https://docs.eigenpal.com/steps/ai/segment.md): Separate a concatenated batch (one big scan) into typed document instances using an LLM. Consumes ai.parse output and a type taxonomy; discovers an unknown number of documents in any order and emits per-document page ranges + text + type, ready for type-specific ai.extract via control.parallel_map.…
- [Classify](https://docs.eigenpal.com/steps/ai/classify.md): Classify a document or text into one of a fixed label set using an LLM. Output exposes the picked label (constrained to the configured names), a coarse confidence, and a short justification. Pair with control.fail to reject documents that match an undesired label.
- [Label Pages](https://docs.eigenpal.com/steps/ai/classify-pages.md): Assign zero or more labels to each page independently (multi-label) using an LLM. Consumes ai.parse output; emits per-page labels and a byLabel map (label -> page indices) that supports NON-contiguous selections. Feed byLabel.<label> straight into ai.vision `pageIndices` to inspect scattered pages o…
- [Inspect Pages (Vision)](https://docs.eigenpal.com/steps/ai/vision.md): Inspect rendered page images with a vision model and return structured JSON matching a schema. The visual counterpart to Extract Data: use it for conclusions that live in the pixels rather than the text (is the document signed? are the photos usable?). Renders PDF, image, or Office/Word inputs; rout…
- [Search Files](https://docs.eigenpal.com/steps/ai/search-files.md): Investigate a ZIP archive with a question and return an answer plus file citations. Read-only: the step opens files in the archive and never writes back. ZIP is the only accepted format in this version. Optionally scope the search to a folder prefix and cap how many files and investigation turns may…
- [Figure Crops](https://docs.eigenpal.com/steps/ai/figure-crops.md): Crop and caption figure elements from a parsed PDF. Renders the source document, crops each kind=figure bbox from ai.parse output, and describes the crop with a vision model. Output is captions plus normalized bboxes for evidence viewers.
- [Set Value](https://docs.eigenpal.com/steps/transform/set.md): Set key-value pairs in the output object
- [Remove Fields](https://docs.eigenpal.com/steps/transform/remove.md): Remove specified fields from an object
- [Combine Data](https://docs.eigenpal.com/steps/transform/combine.md): Merge multiple objects or concatenate arrays
- [Split Data](https://docs.eigenpal.com/steps/transform/split.md): Split a string by delimiter or extract keys from an object
- [Merge Inputs](https://docs.eigenpal.com/steps/transform/merge.md): Merge multiple named inputs into a single output
- [Fill Template](https://docs.eigenpal.com/steps/transform/template.md): Fill a DOCX or XLSX template with data from previous steps.
- [Embed PDF Text](https://docs.eigenpal.com/steps/transform/pdf-embed.md): Embed OCR text layer into scanned PDFs/images to make them searchable
- [Crop Regions](https://docs.eigenpal.com/steps/transform/crop-regions.md): Render a PDF or image and crop normalized bounding boxes into JPEG run output artifacts.
- [Spreadsheet to JSON](https://docs.eigenpal.com/steps/transform/xlsx-to-json.md): Convert an XLS or XLSX spreadsheet to a JSON array of row objects.
- [JSON to XLSX](https://docs.eigenpal.com/steps/transform/json-to-xlsx.md): Convert JSON rows into an XLSX spreadsheet file for download or downstream delivery.
- [Script](https://docs.eigenpal.com/steps/transform/script.md): Execute a TypeScript function in a QuickJS sandbox. Input keys become function parameters in any order, and the required `: R` return-type annotation IS this step's output schema.
- [Text Chunker](https://docs.eigenpal.com/steps/transform/text-chunker.md): Split long text into chunks with regex-anchored boundaries, overlap, and header preservation. Accepts raw text or a parsed-document object; chunks carry source page indexes when pages are provided.
- [Regex Extract](https://docs.eigenpal.com/steps/transform/regex-extract.md): Pull named fields from text via regex patterns (deterministic counterpart to ai.extract). Accepts raw text or a parsed-document object; matches carry `_evidence.pageIndex` when pages are provided.
- [List Archive](https://docs.eigenpal.com/steps/transform/archive-list.md): List files inside a ZIP archive. Returns sorted normalized paths with compressed and uncompressed sizes. Optionally filter by folder prefix or substring, then page with offset and limit.
- [Extract Archive File](https://docs.eigenpal.com/steps/transform/archive-extract.md): Extract exactly one file from a ZIP archive and store it as a run output file. Downstream steps can pass the whole extract output into Parse, Vision, or another archive step.
- [HTTP Request](https://docs.eigenpal.com/steps/action/http.md): Make an HTTP request to an external API
- [Invoke Workflow](https://docs.eigenpal.com/steps/action/invoke-workflow.md): Execute another workflow and return its output
- [Website Reader](https://docs.eigenpal.com/steps/action/website-reader.md): Fetch a webpage and convert content to markdown
- [Send Email](https://docs.eigenpal.com/steps/action/email.md): Send an email with optional attachments via a selected organization email server
- [Condition](https://docs.eigenpal.com/steps/control/if.md): Branch execution based on a condition expression
- [Switch](https://docs.eigenpal.com/steps/control/switch.md): Multi-way routing: resolve an expression and run the first case whose value matches (else default). Cleaner than a nested control.if chain for routing an item to one of N pipelines by a discriminator field like a document type.
- [For Each](https://docs.eigenpal.com/steps/control/foreach.md): Loop over an array and execute steps for each item
- [Parallel Map](https://docs.eigenpal.com/steps/control/parallel_map.md): Iterate over an array with concurrent execution up to a limit
- [Parallel](https://docs.eigenpal.com/steps/control/parallel.md): Execute multiple branches concurrently
- [Wait](https://docs.eigenpal.com/steps/control/wait.md): Pause workflow execution for a specified duration
- [Human Review](https://docs.eigenpal.com/steps/control/human_review.md): Pause the run for selective field confirmation or correction, then continue with the reviewed data.
- [Fail](https://docs.eigenpal.com/steps/control/fail.md): Terminate the workflow with a typed status code + message. With an optional condition, only fails when the condition is truthy; otherwise always fails when reached. Pair with ai.classify or any prior step to fail fast on bad inputs.
- [Your first workflow](https://docs.eigenpal.com/guides/first-workflow.md): Scaffold, push, and run a workflow from the CLI in a few minutes.
- [Your first agent](https://docs.eigenpal.com/guides/first-agent.md): Clone, edit, release, run, and inspect a Git-backed agent.
- [Eval-first development](https://docs.eigenpal.com/guides/eval-first-development.md): Build a workflow, define evaluators, run experiments, and turn failures into regression tests.
- [Build a dataset](https://docs.eigenpal.com/guides/build-a-dataset.md): Create example inputs and expected outputs, then push them as a portable folder archive.
- [Evaluate a workflow](https://docs.eigenpal.com/guides/evaluate-workflow.md): Define evaluators, run an experiment over a dataset, and compare versions.
- [Deploy to production](https://docs.eigenpal.com/guides/deploy-to-production.md): Take a workflow from a passing evaluation to full automation, ramping production traffic with sampled human review.
- [Upload files](https://docs.eigenpal.com/guides/upload-files.md): Upload reusable files directly or through the API, then reference them from runs and datasets.
- [Receive run events with webhooks](https://docs.eigenpal.com/guides/outbound-webhooks.md): Send signed run lifecycle events from EigenPal to your application.
- [Manage email servers](https://docs.eigenpal.com/guides/email-servers.md): Configure organization Resend and SMTP servers for the Send Email workflow step.
- [Review production runs](https://docs.eigenpal.com/guides/review-runs.md): Review runs, track fixes with status, rerun the same inputs, and monitor quality over time.
- [Execution tags](https://docs.eigenpal.com/guides/execution-tags.md): Correlate runs with request IDs and find their outputs later.
- [Optimize cost](https://docs.eigenpal.com/guides/optimize-cost.md): Fix the quality bar with evaluations, then pick the cheapest model configuration that still passes.
- [Install the EigenPal skill](https://docs.eigenpal.com/guides/skill-install.md): Add the EigenPal CLI skill to Claude Code, Cursor, and other AI coding tools.
- [Changelog](https://docs.eigenpal.com/changelog.md): Production release history for the EigenPal platform.
- [CLI overview](https://docs.eigenpal.com/cli/overview.md): Install the EigenPal CLI, authenticate, and manage workflows, datasets, runs, agents, and email servers.
- [Status](https://docs.eigenpal.com/cli/status.md): One-shot dashboard: server, active tenant, user, key id, workflow count. Pair with --json for scripting.
- [Init](https://docs.eigenpal.com/cli/init.md): Scaffold a new workflow project. Without [name], scaffolds into the current directory using the cwd basename as the workflow name. With [name], creates ./<name>/ and uses that as the slug. The flat layout matches what run workflows.<slug> already discovers, no manual file moves.
- [Auth](https://docs.eigenpal.com/cli/auth.md): Manage authentication. Credentials live in ~/.config/eigenpal/credentials.json as named profiles. Switch tenants with auth use <name> or set EIGENPAL_PROFILE=<name> for one shell.
- [Models](https://docs.eigenpal.com/cli/models.md): Inspect models and providers configured for the current tenant environment (text, vision, OCR).
- [Email servers](https://docs.eigenpal.com/cli/email-servers.md): Manage outbound email servers (ems\_…): list, inspect, create, update, delete, and send a real test message.
- [Workflow](https://docs.eigenpal.com/cli/workflow.md): Manage workflows: push, pull, and evaluate.
- [Agents](https://docs.eigenpal.com/cli/agents.md): Manage EigenPal agents: Git source, datasets, experiments, sessions, and releases.
- [Run](https://docs.eigenpal.com/cli/run.md): Start a workflow or agent run, e.g. workflows.extract-invoice.
- [Rerun](https://docs.eigenpal.com/cli/rerun.md): Create a new run from a previous run's stored input snapshot.
- [Runs](https://docs.eigenpal.com/cli/runs.md): Inspect, watch, and manage workflow, agent, and eval runs.
- [Reviews](https://docs.eigenpal.com/cli/reviews.md): Work the live human-review queue for paused runs: list pending tasks, confirm fields, approve, reject, and download attachments.
- [Skill](https://docs.eigenpal.com/cli/skill.md): Install / uninstall / list the EigenPal agent skill in your project (Claude Code, Cursor, Codex, Gemini CLI, Antigravity, OpenCode, Pi, Windsurf, GitHub Copilot).
- [Docs](https://docs.eigenpal.com/cli/docs.md): List, search, and read the complete release-matched EigenPal documentation offline.
- [SDKs](https://docs.eigenpal.com/sdks/overview.md): Official TypeScript and Python SDKs for triggering EigenPal workflows and agents from your code.
- [TypeScript SDK](https://docs.eigenpal.com/sdks/typescript.md): Trigger EigenPal workflows and agents from TypeScript with @eigenpal/sdk.
- [Reference](https://docs.eigenpal.com/sdks/typescript/reference.md)
- [Configuration](https://docs.eigenpal.com/sdks/typescript/configuration.md)
- [Automations](https://docs.eigenpal.com/sdks/typescript/automations.md)
- [Evaluation](https://docs.eigenpal.com/sdks/typescript/evaluation.md)
- [Runs](https://docs.eigenpal.com/sdks/typescript/runs.md)
- [Files](https://docs.eigenpal.com/sdks/typescript/files.md)
- [Errors](https://docs.eigenpal.com/sdks/typescript/errors.md)
- [Python SDK](https://docs.eigenpal.com/sdks/python.md): Trigger EigenPal workflows and agents from Python with the eigenpal package.
- [Reference](https://docs.eigenpal.com/sdks/python/reference.md)
- [Configuration](https://docs.eigenpal.com/sdks/python/configuration.md)
- [Automations](https://docs.eigenpal.com/sdks/python/automations.md)
- [Evaluation](https://docs.eigenpal.com/sdks/python/evaluation.md)
- [Runs](https://docs.eigenpal.com/sdks/python/runs.md)
- [Files](https://docs.eigenpal.com/sdks/python/files.md)
- [Errors](https://docs.eigenpal.com/sdks/python/errors.md)
- [Check API key identity](https://docs.eigenpal.com/api-reference/metadata/check-api-key-identity.md): Return the tenant, user, API key, and scope represented by the current API key.
- [List configured models](https://docs.eigenpal.com/api-reference/models/list-configured-models.md): List text, vision, and OCR models configured for this tenant's environment from the workspace model catalog. This is a cheap read-only inventory: it does not call providers. `health` is `configured` or `unconfigured` from local credentials, never a live probe. Secrets and provider endpoints are neve…
- [List automations](https://docs.eigenpal.com/api-reference/automations/list-automations.md): Returns workflows and agents through one runnable automation collection. Use `type` to narrow to workflows or agents, `search` to find automations by slug, name, or description, and `folderId` to list YAML workflows in a folder (`null` for root).
- [Get automation](https://docs.eigenpal.com/api-reference/automations/get-automation.md): Get one runnable workflow or agent automation by id or typed alias.
- [Move workflow automation](https://docs.eigenpal.com/api-reference/automations/move-workflow-automation.md): Move a YAML workflow between organizing folders. `folderPath` auto-creates missing workflow folders; empty or `/` files the workflow at root. Agent automations have no database folder model and are rejected. Identifiers match GET (workflow id, automation id, or `workflows.<slug>`).
- [Delete automation](https://docs.eigenpal.com/api-reference/automations/delete-automation.md): Delete a workflow or agent automation using the same cleanup as the dashboard. Workflows archive the automations registry parent and keep execution history. Agents delete the agent row, archive the registry parent, and best-effort-delete agent storage. Identifiers match GET.
- [List automation versions](https://docs.eigenpal.com/api-reference/automations/list-automation-versions.md): List versions for a workflow or agent automation. YAML workflow lists include tagged releases plus the current untagged snapshot when HEAD is untagged (for example after restore), so the current version is always present. Agent lists remain Git release tags.
- [Create a workflow version](https://docs.eigenpal.com/api-reference/automations/create-a-workflow-version.md): Create a tagged YAML workflow candidate from validated YAML or by copying an existing snapshot (`historyId`). Provide exactly one of `yaml` or `historyId`. Copy creates a new tagged row and leaves the source tag unchanged; it does not retag the original. Defaults to making the new version current. S…
- [Restore a workflow version](https://docs.eigenpal.com/api-reference/automations/restore-a-workflow-version.md): Restore a YAML workflow automation by copying a previous snapshot into a new untagged current version. The source tag is left unchanged; the new HEAD appears in subsequent version lists as the untagged current row and cannot be promoted until you create a tagged copy. The JSON body may be `{}`; `mes…
- [Promote a workflow version](https://docs.eigenpal.com/api-reference/automations/promote-a-workflow-version.md): Make an existing tagged YAML workflow candidate current without creating another history row. Only tagged version rows can be promoted; untagged snapshots (including restore HEAD) and missing ids return 404. Agent automations return 400. Requires a Bearer API key or a dashboard session.
- [Get automation triggers](https://docs.eigenpal.com/api-reference/automations/get-automation-triggers.md): Read trigger state for a workflow or agent automation. Trigger mutation is not public v1.
- [Sync automation from latest Git release](https://docs.eigenpal.com/api-reference/automations/sync-automation-from-latest-git-release.md): Reconciles automation registry metadata and trigger projections from the latest Git source release. This operation is idempotent for unchanged source state: repeated calls against the same latest release leave the same automation registry state and may repeat the same warnings. Requires a Bearer API…
- [List folders](https://docs.eigenpal.com/api-reference/folders/list-folders.md): List folders in one tree. `type` is required (`workflow` or `template`). Pass `tree=true` for the full nested tree with counts; otherwise list direct children of `parentId` (`null` for root). Deleting a folder later unfiles contained workflows or templates and does not delete them.
- [Create folder](https://docs.eigenpal.com/api-reference/folders/create-folder.md): Create a folder in the workflow or template tree. `type` is required. Missing parents 404; a same-named sibling at that location returns 409. Nested agent directories in Git are source organization only and are not folders.
- [Get folder](https://docs.eigenpal.com/api-reference/folders/get-folder.md): Get one workflow or template folder by id.
- [Update folder](https://docs.eigenpal.com/api-reference/folders/update-folder.md): Rename or reparent a folder. Moving a folder onto itself, into a descendant, or into the other tree is rejected. A same-named sibling at the destination returns 409.
- [Delete folder](https://docs.eigenpal.com/api-reference/folders/delete-folder.md): Delete a folder and cascade-delete child folders. Workflows and templates in the tree are unfiled, not deleted.
- [Start a run](https://docs.eigenpal.com/api-reference/runs/start-a-run.md): Start a run. Send JSON or multipart/form-data.
- [List runs](https://docs.eigenpal.com/api-reference/runs/list-runs.md): List workflow and agent runs with cursor pagination.
- [Get a run](https://docs.eigenpal.com/api-reference/runs/get-a-run.md): Fetch one run by id. By default this returns core metadata plus terminal output/error fields. Pass `?expand=input,usage,execution,debug` to include detailed sub-objects; `expand=execution` is also where embedded review and expected artifacts appear.
- [Cancel run](https://docs.eigenpal.com/api-reference/runs/cancel-run.md): Cancel a queued run or request cancellation of an in-flight run.
- [Retry run](https://docs.eigenpal.com/api-reference/runs/retry-run.md): Start a new run using the source run input. By default the retry uses the latest automation version; pass `version=original` to pin the same source version as the original run.
- [Get run usage](https://docs.eigenpal.com/api-reference/runs/get-run-usage.md): Get token, credit, duration, and execution usage for a run.
- [List run artifacts](https://docs.eigenpal.com/api-reference/runs/list-run-artifacts.md): Returns a JSON list of downloadable artifact paths for a run. Pass `zip=1` to switch the response to a ZIP download containing output files.
- [Download run artifact](https://docs.eigenpal.com/api-reference/runs/download-run-artifact.md): Download one artifact by path. Cloud deployments may redirect to a short-lived storage URL for large files.
- [List run steps](https://docs.eigenpal.com/api-reference/runs/list-run-steps.md): List slim workflow step rows for a run (no input/output payloads) plus the total count. Fetch a single step for full payloads.
- [Get a run step](https://docs.eigenpal.com/api-reference/runs/get-a-run-step.md): Return the full single step execution (including input/output) for on-demand inspection. The step must belong to this run or a child invoke-workflow run.
- [List run events](https://docs.eigenpal.com/api-reference/runs/list-run-events.md): List a stable chronological lifecycle timeline for a run.
- [Get run trace](https://docs.eigenpal.com/api-reference/runs/get-run-trace.md): Return low-level execution trace events for debugging one run. Workflow runs expose observability phases or step records; agent runs expose parsed trace.jsonl events. The shape is intentionally extensible, but common fields are documented.
- [Webhook receiver reference](https://docs.eigenpal.com/api-reference/webhooks.md): Receive, verify, and process EigenPal webhook events.
- [Get automation review health](https://docs.eigenpal.com/api-reference/reviews/get-automation-review-health.md): Aggregates reviewed correctness, review coverage, bucketed counts, and rolling-window confidence for one automation. Prefer this endpoint for single-automation monitoring dashboards.
- [Get run review](https://docs.eigenpal.com/api-reference/reviews/get-run-review.md): Returns review metadata and corrections for a run. Corrected files are listed at GET /runs/{id}/reviews/expected; embed review + expected artifacts with GET /runs/{id}?expand=execution.
- [Update run review](https://docs.eigenpal.com/api-reference/reviews/update-run-review.md): Create or replace review metadata for a run.
- [Clear run review](https://docs.eigenpal.com/api-reference/reviews/clear-run-review.md): Deletes review metadata, corrections, and corrected files for the run.
- [List corrected files](https://docs.eigenpal.com/api-reference/reviews/list-corrected-files.md): Returns corrected artifact files attached to the run review. Review metadata and corrected JSON output live at GET /runs/{id}/reviews.
- [Add corrected file](https://docs.eigenpal.com/api-reference/reviews/add-corrected-file.md): Attach one corrected file to a run review. Send multipart/form-data with `file` and optional `name` to upload a local file, or JSON with `outputFileName` and optional `expectedName` to copy an existing run output file.
- [Download corrected artifact file](https://docs.eigenpal.com/api-reference/reviews/download-corrected-artifact-file.md): Downloads one corrected artifact file attached to the run review. Use the `filename` returned by the corrected-output collection endpoint.
- [Rename corrected artifact file](https://docs.eigenpal.com/api-reference/reviews/rename-corrected-artifact-file.md): Renames one corrected artifact file attached to the run review.
- [Delete corrected artifact file](https://docs.eigenpal.com/api-reference/reviews/delete-corrected-artifact-file.md): Deletes one corrected artifact file attached to the run review.
- [Promote run to example](https://docs.eigenpal.com/api-reference/reviews/promote-run-to-example.md): Turn a reviewed run into a dataset example. The new example uses the run input and any corrected output/files stored through the review endpoints.
- [Upload file](https://docs.eigenpal.com/api-reference/files/upload-file.md): Upload a reusable file that can later be referenced by run inputs or dataset examples.
- [Prepare file upload](https://docs.eigenpal.com/api-reference/files/prepare-file-upload.md): Negotiate HTTP multipart for small bodies, a short-lived signed PUT under the single-object ceiling, or storage-direct multipart (presigned-multipart) for larger files when storage supports MPU. The response transport is authoritative; clients must not guess from file size alone.
- [Get file upload session](https://docs.eigenpal.com/api-reference/files/get-file-upload-session.md): Return session status for resume. Multipart sessions include authoritative uploaded parts from storage, not client-reported ETags.
- [List uploaded multipart parts](https://docs.eigenpal.com/api-reference/files/list-uploaded-multipart-parts.md): List authoritative uploaded parts from storage for resume. Do not trust client-only ETags.
- [Presign one multipart upload part](https://docs.eigenpal.com/api-reference/files/presign-one-multipart-upload-part.md): Mint a short-lived signed UploadPart URL for one validated part. Part URLs are not issued at session create.
- [Complete file upload](https://docs.eigenpal.com/api-reference/files/complete-file-upload.md): Verify a storage-direct pending object and promote it into a reusable file. Safe to retry.
- [Abort file upload](https://docs.eigenpal.com/api-reference/files/abort-file-upload.md): Abort a pending storage-direct upload. Multipart sessions call AbortMultipartUpload; pending PUT objects are deleted. Completed canonical files are never deleted.
- [Get file metadata](https://docs.eigenpal.com/api-reference/files/get-file-metadata.md): Get metadata for a reusable uploaded file.
- [Download file content](https://docs.eigenpal.com/api-reference/files/download-file-content.md): Download bytes for a reusable uploaded file. Direct-enabled deployments redirect responses above their configured body limit to a short-lived signed storage URL; multipart-only/on-prem deployments stream through the API.
- [Delete file](https://docs.eigenpal.com/api-reference/files/delete-file.md): Delete a reusable uploaded file. Past runs that referenced it stay readable until those runs are deleted.
- [List templates](https://docs.eigenpal.com/api-reference/templates/list-templates.md): List tenant-scoped DOCX and XLSX template resources.
- [Upload template](https://docs.eigenpal.com/api-reference/templates/upload-template.md): Create a stable `tmpl_…` resource and its first immutable content revision from a reusable `fileId`. Public SDK helpers `create(file)` and `createFromFileId(fileId)` upload through the Files API when needed, then send this JSON body. Generated clients send `{ fileId }` JSON only. The HTTP route stil…
- [Inspect template](https://docs.eigenpal.com/api-reference/templates/inspect-template.md): Get template metadata, checksum, discovered tokens, grammar capabilities, and current immutable revision. Storage keys are never exposed.
- [Download template content](https://docs.eigenpal.com/api-reference/templates/download-template-content.md): Download current bytes while the logical template exists, or a specific immutable revision using `revisionId`. Pinned revision downloads remain available after logical template deletion. Large objects may 302 to a short-lived signed storage URL.
- [Create template revision](https://docs.eigenpal.com/api-reference/templates/create-template-revision.md): Append an immutable revision and advance the logical template pointer from a reusable `fileId`. Public SDK helpers `replace(file)` and `replaceFromFileId(fileId)` upload through the Files API when needed, then send this JSON body. Generated clients send `{ fileId }` JSON only. The HTTP route still a…
- [Finalize or hard-clean a staged template](https://docs.eigenpal.com/api-reference/templates/finalize-or-hard-clean-a-staged-template.md): Consume the one-time cleanupProof issued on a staged create. `finalize` makes the template live so later deletion keeps pinned revisions. `cleanup` hard-removes only unpublished resources from that staging attempt. Normal DELETE is unchanged and never takes this path.
- [Delete template](https://docs.eigenpal.com/api-reference/templates/delete-template.md): Delete the mutable logical template. Immutable revisions are retained so workflows pinned with `templateRevisionId` continue to execute; unpinned workflows can no longer resolve the deleted `tmpl_…` id.
- [List email servers](https://docs.eigenpal.com/api-reference/email-servers/list-email-servers.md): List outbound email servers for the current workspace. Secrets are never returned; each server includes configuration flags instead of credentials.
- [Create email server](https://docs.eigenpal.com/api-reference/email-servers/create-email-server.md): Create an outbound email server. Secrets are encrypted at rest and never returned. Names must be unique among live servers in the workspace.
- [Get email server](https://docs.eigenpal.com/api-reference/email-servers/get-email-server.md): Inspect a stored outbound email server. Cross-tenant and deleted ids are indistinguishable from missing. Secrets are never returned.
- [Update email server](https://docs.eigenpal.com/api-reference/email-servers/update-email-server.md): Rename, enable/disable, or replace transport configuration. Omitted secrets are retained only when the existing transport is compatible and, for SMTP, the host and username are unchanged. Changing transport or SMTP destination requires a complete valid target; a new password or explicit auth clearin…
- [Delete email server](https://docs.eigenpal.com/api-reference/email-servers/delete-email-server.md): Soft-delete an outbound email server. Deleted ids are indistinguishable from missing.
- [Test email server](https://docs.eigenpal.com/api-reference/email-servers/test-email-server.md): Send a concise connectivity email through the selected stored server. Disabled servers conflict. Provider failures are returned as a generic outcome and never include credentials or the test recipient.
- [Export automation dataset](https://docs.eigenpal.com/api-reference/evaluation/export-automation-dataset.md): Download the automation dataset as a ZIP archive. The archive uses the examples/<name>/input and examples/<name>/expected folder convention, so it can be re-imported into another automation or environment. Cloud deployments may redirect large archives to a short-lived storage URL.
- [Import automation dataset](https://docs.eigenpal.com/api-reference/evaluation/import-automation-dataset.md): Import a dataset ZIP archive using the examples/<name>/input and examples/<name>/expected folder convention. Use `mode=append` for additive imports or `mode=replace` to replace the dataset.
- [List dataset examples](https://docs.eigenpal.com/api-reference/evaluation/list-dataset-examples.md): List dataset examples for one automation. Examples contain input, expected output, expected files, metadata, and optional overrides used by evaluation runs. Pass `include=metadata` to return ids, names, metadata, and expected file refs without loading input or expected JSON.
- [Create dataset example](https://docs.eigenpal.com/api-reference/evaluation/create-dataset-example.md): Create one dataset example from JSON fields. Use dataset import for archive-based uploads and file-bearing examples.
- [Get dataset example](https://docs.eigenpal.com/api-reference/evaluation/get-dataset-example.md): Fetch one dataset example, including input, expected output, expected files, metadata, and overrides.
- [Update dataset example](https://docs.eigenpal.com/api-reference/evaluation/update-dataset-example.md): Partially update a dataset example. Omitted fields are preserved; pass null for nullable fields to clear them.
- [Delete dataset example](https://docs.eigenpal.com/api-reference/evaluation/delete-dataset-example.md): Delete one dataset example from the automation dataset. This removes the example from future experiments.
- [Run dataset example](https://docs.eigenpal.com/api-reference/evaluation/run-dataset-example.md): Start an asynchronous run using the input from one dataset example. Poll `GET /api/v1/runs/{id}` for completion and use run scores or review endpoints to review the result.
- [List input files](https://docs.eigenpal.com/api-reference/evaluation/list-input-files.md): List files stored under the input folder for one automation dataset example.
- [Upload input files](https://docs.eigenpal.com/api-reference/evaluation/upload-input-files.md): Upload one or more files into the input folder for an automation dataset example. Use `$file` references such as `input/invoice.pdf` from the example input JSON to consume them.
- [Download input file](https://docs.eigenpal.com/api-reference/evaluation/download-input-file.md): Download one file from an automation dataset example input folder.
- [Rename input file](https://docs.eigenpal.com/api-reference/evaluation/rename-input-file.md): Rename one input file attached to an automation dataset example. The parent folder is preserved.
- [Delete input file](https://docs.eigenpal.com/api-reference/evaluation/delete-input-file.md): Delete one file from an automation dataset example input folder.
- [List expected files](https://docs.eigenpal.com/api-reference/evaluation/list-expected-files.md): List files stored under the expected folder for one automation dataset example.
- [Upload expected files](https://docs.eigenpal.com/api-reference/evaluation/upload-expected-files.md): Upload one or more files into the expected folder for an automation dataset example. Use `$file` references such as `expected/result.pdf` from expected JSON to compare file outputs.
- [Download expected dataset file](https://docs.eigenpal.com/api-reference/evaluation/download-expected-dataset-file.md): Download one expected file attached to an automation dataset example.
- [Rename expected file](https://docs.eigenpal.com/api-reference/evaluation/rename-expected-file.md): Rename one expected file attached to an automation dataset example. The parent folder is preserved.
- [Delete expected file](https://docs.eigenpal.com/api-reference/evaluation/delete-expected-file.md): Delete one file from an automation dataset example expected folder.
- [Get evaluators](https://docs.eigenpal.com/api-reference/evaluation/get-evaluators.md): Fetch the evaluator configuration for an automation. Evaluators produce automated `score` results, which are separate from human review verdicts.
- [Replace evaluators](https://docs.eigenpal.com/api-reference/evaluation/replace-evaluators.md): Replace the evaluator YAML for an automation. The submitted YAML is validated before it becomes the source for future experiment scores.
- [List experiments](https://docs.eigenpal.com/api-reference/evaluation/list-experiments.md): List experiment batches for one automation. Each experiment runs selected dataset examples and records automated evaluator scores.
- [Create experiment](https://docs.eigenpal.com/api-reference/evaluation/create-experiment.md): Start an asynchronous experiment batch for one automation. Omit `examples` to run the full dataset, or pass specific example ids to run a subset.
- [Create automation experiment with NDJSON progress](https://docs.eigenpal.com/api-reference/evaluation/create-automation-experiment-with-ndjson-progress.md): Starts a batch eval experiment for workflow or agent automations and streams per-run completion events as NDJSON.
- [Get experiment](https://docs.eigenpal.com/api-reference/evaluation/get-experiment.md): Fetch one experiment batch with its run summaries and evaluator results grouped by run id.
- [Cancel experiment](https://docs.eigenpal.com/api-reference/evaluation/cancel-experiment.md): Request cancellation for an experiment batch. Already-completed runs remain recorded; queued or running work is cancelled when possible.
- [Export experiment eval results](https://docs.eigenpal.com/api-reference/evaluation/export-experiment-eval-results.md): Download eval result rows for a single experiment batch as CSV or JSON.
- [Export all experiment eval results](https://docs.eigenpal.com/api-reference/evaluation/export-all-experiment-eval-results.md): Download every eval result row for an automation as CSV or JSON.
- [Resolve experiment by id](https://docs.eigenpal.com/api-reference/evaluation/resolve-experiment-by-id.md): Returns the owning automation for an experiment batch id. Used when callers only know the experiment id.
- [List run evaluator scores](https://docs.eigenpal.com/api-reference/evaluation/list-run-evaluator-scores.md): List automated evaluator results for one run. Use `score` for evaluator output and run reviews for human verdicts.
- [Create dataset review request](https://docs.eigenpal.com/api-reference/evaluation/create-dataset-review-request.md): Snapshot selected dataset examples into a review request. Include instructions, focusFields to highlight, and ignoredFields for paths reviewers can skip. Optional itemNotes become commented events on those examples.
- [Get dataset review request](https://docs.eigenpal.com/api-reference/evaluation/get-dataset-review-request.md): Fetch one dataset review request with snapshotted items, progress, focus fields, ignored fields, and events.
- [Download dataset review item file](https://docs.eigenpal.com/api-reference/evaluation/download-dataset-review-item-file.md): Download one input file (default) or expected-output file (`?kind=expected`) belonging to a review item. Authorized by `dataset_review:read` and scoped to files referenced in the item snapshot, so reviewers without workflow read can still preview documents. Expected files resolve to reviewer-correct…
- [List dataset review requests](https://docs.eigenpal.com/api-reference/evaluation/list-dataset-review-requests.md): List dataset review requests for one automation, including per-request progress.
- [List dataset review events](https://docs.eigenpal.com/api-reference/evaluation/list-dataset-review-events.md): List audit events for one dataset review request.
- [List dataset review items](https://docs.eigenpal.com/api-reference/evaluation/list-dataset-review-items.md): List items for one dataset review request, optionally filtered by item status.
- [Update dataset review request](https://docs.eigenpal.com/api-reference/evaluation/update-dataset-review-request.md): Update review metadata or lifecycle status. Set status to `closed` when review is finished; a closed review accepts only a bare reopen back to `open`. Dataset reconciliation stays manual.
- [Update dataset review item](https://docs.eigenpal.com/api-reference/evaluation/update-dataset-review-item.md): Approve, edit, remove, reopen, comment, or record a field-decision or file-decision on one review item while the parent request is draft, open, or paused. Pass `expectedUpdatedAt` from the item the client last observed. Pass `fieldPath` with `action: comment` or `action: field-decision`, `filePath`…
- [List dataset review requests across the tenant](https://docs.eigenpal.com/api-reference/evaluation/list-dataset-review-requests-across-the-tenant.md): Tenant-wide inbox of dataset review requests with progress and automation display names. Gated by `dataset_review:read` (reviewers allowed; does not require `workflow:read`).

## OpenAPI Specs

- [openapi](/api-reference/openapi.json)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.