Skip to main content

Commands

Core

CommandDescription
eigenpal workflow list|ls [options]List workflows the caller can read.
eigenpal workflow pull [options] <workflow-id>Download the YAML definition of the workflow at its current version.
eigenpal workflow push [options]Create or update a workflow from a YAML file.
eigenpal workflow move [options] <workflow-id>Move a workflow to a folder path, creating folders as needed
eigenpal workflow validate [options] [path]Local-only validation. Without [path]: runs the templated three-way check (./workflow.yaml + ./evaluators.yaml + ./dataset/). With [path] pointing at a YAML file: validates just that workflow.yaml. For per-noun targeting use evaluators validate or dataset validate.
eigenpal workflow clear-local [options] [examples...]Delete local execution artifacts under ./dataset/examples/. Keeps the latest run per example by default.

Evaluators

CommandDescription
eigenpal workflow evaluators pull [options] <workflow-id>Download the workflow’s evaluators YAML.
eigenpal workflow evaluators push [options] <workflow-id>Overwrite the workflow’s evaluator config from a YAML file.
eigenpal workflow evaluators validate [options] [path]Validate an evaluators YAML file against the EvalConfig schema. Defaults to ./evaluators.yaml.

Dataset

CommandDescription
eigenpal workflow dataset list|ls [options] <workflow-id>List eval examples for the workflow.
eigenpal workflow dataset pull [options] <workflow-id>Download the workflow’s dataset as a ZIP archive.
eigenpal workflow dataset push [options] <workflow-id>Replace or extend the workflow’s dataset from a ZIP or folder.
eigenpal workflow dataset example create [options] <workflow-id>Create one eval example without re-uploading the dataset.
eigenpal workflow dataset example update [options] <workflow-id> <exampleId>Patch one eval example. Omitted flags are left alone.
eigenpal workflow dataset example delete [options] <workflow-id> <exampleId>Delete one eval example by id. Non-TTY shells require —yes.
eigenpal workflow dataset example get [options] <workflow-id> <exampleId>Fetch one eval example with full triggerInput, expectedOutput, and metadata.
eigenpal workflow dataset validate [options] [path]Validate a dataset folder against the examples/<name>/{input,expected,meta} convention. Defaults to ./dataset/.

Experiment

CommandDescription
eigenpal workflow experiment|exp list|ls [options] <workflow-id>List executions for the workflow, newest first.
eigenpal workflow experiment|exp run [options] <workflow-id>Start a batch eval against the workflow’s dataset.
eigenpal workflow experiment|exp status [options] <workflow-id> <batchId>Aggregate progress for a batch by batchId.
eigenpal workflow experiment|exp cancel [options] <workflow-id> <batchId>Cancel every execution in a batch. Idempotent.
eigenpal workflow experiment|exp results [options] <workflow-id> [batchId]Download eval results in CSV or JSON.
eigenpal workflow experiment|exp compare|diff [options] <batchIdA> <batchIdB>Diff eval scores between two experiment batches.
eigenpal workflow experiment|exp watch [options] <workflow-id> <batchId>Poll until terminal, then auto-pull results, replaces status --watch + results --out.

Versions

CommandDescription
eigenpal workflow versions list|ls [options] <workflow-id>List historical workflow versions, newest first.
eigenpal workflow versions restore [options] <workflow-id> <versionId>Restore the workflow to a previous version.

Step-type

CommandDescription
eigenpal workflow step-type list|ls [options]List every step type the deployment supports.
eigenpal workflow step-type get [options] <type>Return the full schema and behavioral docs for one step type.

Evaluator-type

CommandDescription
eigenpal workflow evaluator-type list|ls [options]List every evaluator type with a one-line description.
eigenpal workflow evaluator-type get [options] <type>Fetch the JSON Schema for one evaluator type. Pipe through jq to inspect specific fields.

Step

CommandDescription
eigenpal workflow step exec [options] <type>DISABLED, local mimic runners removed pending server-side redesign (EIG-104). Use run or workflow experiment run instead.

Details

eigenpal workflow list|ls [options]

List workflows the caller can read.

Options

FlagRequiredDefaultDescription
--search <q>noFilter by name
--limit <n>no50Page size
--offset <n>no0Page offset
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow pull [options] <workflow-id>

Download the YAML definition of the workflow at its current version.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--out <path>noWrite YAML to file instead of stdout
--base-url <url>noServer base URL

eigenpal workflow push [options]

Create or update a workflow from a YAML file.

Options

FlagRequiredDefaultDescription
--file <yaml>noPath to YAML file
--workflow-id <id>noUpdate existing workflow (default: create new)
--bump <level>noAuto-bump from the server’s current version: patch | minor | major. Mutually exclusive with --version and with a top-level version: in the YAML.
--set-version <semver>noPush at this exact semver (e.g. 1.4.0). Mutually exclusive with --bump and with a top-level version: in the YAML. (Named --set-version to avoid the global -v, --version flag.)
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow move [options] <workflow-id>

Move a workflow to a folder path, creating folders as needed

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--folder <path>yesTarget folder path (/ for root)
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow validate [options] [path]

Local-only validation. Without [path]: runs the templated three-way check (./workflow.yaml + ./evaluators.yaml + ./dataset/). With [path] pointing at a YAML file: validates just that workflow.yaml. For per-noun targeting use evaluators validate or dataset validate.

Arguments

NameRequiredVariadicDescription
pathnono

Options

FlagRequiredDefaultDescription
--dir <path>noProject root (defaults to cwd; resolves the three default paths from here)

eigenpal workflow clear-local [options] [examples...]

Delete local execution artifacts under ./dataset/examples/. Keeps the latest run per example by default.

Arguments

NameRequiredVariadicDescription
examplesnoyes

Options

FlagRequiredDefaultDescription
--dir <dir>noLocal eigenpal directory
--allnofalseRemove all artifacts, including the latest kept by default

eigenpal workflow evaluators pull [options] <workflow-id>

Download the workflow’s evaluators YAML.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--out <path>noWrite YAML to file instead of stdout
--base-url <url>noServer base URL

eigenpal workflow evaluators push [options] <workflow-id>

Overwrite the workflow’s evaluator config from a YAML file.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--file <yaml>yesPath to evaluators YAML file
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow evaluators validate [options] [path]

Validate an evaluators YAML file against the EvalConfig schema. Defaults to ./evaluators.yaml.

Arguments

NameRequiredVariadicDescription
pathnono

eigenpal workflow dataset list|ls [options] <workflow-id>

List eval examples for the workflow.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--limit <n>no100Page size
--offset <n>no0Page offset
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow dataset pull [options] <workflow-id>

Download the workflow’s dataset as a ZIP archive.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--out <zip>noWrite the dataset ZIP to this path. When omitted, the binary streams to stdout.
--example-id <id>no[]Export only this example (repeatable). When omitted, the whole dataset is exported.
--base-url <url>noServer base URL

eigenpal workflow dataset push [options] <workflow-id>

Replace or extend the workflow’s dataset from a ZIP or folder.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--file <path>yesPath to a dataset ZIP file or a dataset/ folder (folder is zipped in memory before upload)
--mode <append|replace>no"append"Import mode (default: append)
--yesnofalseSkip the destructive confirmation prompt for —mode replace (use in CI)
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow dataset example create [options] <workflow-id>

Create one eval example without re-uploading the dataset.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--name <name>yesExample name (1–64 chars)
--input-json <json>noTrigger input as a JSON literal
--input-file <path>noTrigger input from a JSON file (or - for stdin)
--expected-json <json>noExpected output as a JSON literal
--expected-file <path>noExpected output from a JSON file (or - for stdin)
--annotation <text>noFree-form annotation
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow dataset example update [options] <workflow-id> <exampleId>

Patch one eval example. Omitted flags are left alone.

Arguments

NameRequiredVariadicDescription
workflow-idyesno
exampleIdyesno

Options

FlagRequiredDefaultDescription
--name <name>noRename the example (1–64 chars)
--input-json <json>noReplace trigger input with this JSON literal
--input-file <path>noReplace trigger input from a JSON file (or - for stdin)
--expected-json <json>noReplace expected output with this JSON literal
--expected-file <path>noReplace expected output from a JSON file (or - for stdin)
--annotation <text>noReplace annotation; pass empty string to clear
--row-order <n>noReorder the row (0-based)
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow dataset example delete [options] <workflow-id> <exampleId>

Delete one eval example by id. Non-TTY shells require —yes.

Arguments

NameRequiredVariadicDescription
workflow-idyesno
exampleIdyesno

Options

FlagRequiredDefaultDescription
--yesnofalseRequired for non-TTY shells; explicit acknowledgment that this is destructive
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow dataset example get [options] <workflow-id> <exampleId>

Fetch one eval example with full triggerInput, expectedOutput, and metadata.

Arguments

NameRequiredVariadicDescription
workflow-idyesno
exampleIdyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow dataset validate [options] [path]

Validate a dataset folder against the examples/<name>/{input,expected,meta} convention. Defaults to ./dataset/.

Arguments

NameRequiredVariadicDescription
pathnono

eigenpal workflow experiment|exp list|ls [options] <workflow-id>

List executions for the workflow, newest first.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--batch-id <id>noFilter by batch
--limit <n>no50Page size
--offset <n>no0Page offset
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow experiment|exp run [options] <workflow-id>

Start a batch eval against the workflow’s dataset.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--example-id <id>no[]Run only this example (repeatable)
--waitnofalsePoll until terminal; non-zero exit on passRate < 1.0
--interval <n>no10Polling interval in seconds (default 10)
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow experiment|exp status [options] <workflow-id> <batchId>

Aggregate progress for a batch by batchId.

Arguments

NameRequiredVariadicDescription
workflow-idyesno
batchIdyesno

Options

FlagRequiredDefaultDescription
--watchnofalsePoll until every execution reaches a terminal state (completed/failed/cancelled/rejected), then exit
--shortnofalseSingle-line plain-text summary on stdout (e.g. 6/6 done failed=0 cancelled=0 rejected=0). Pipe-friendly for monitoring loops; mutually exclusive with —watch.
--interval <seconds>no5Poll interval in seconds when —watch is set (default 5)
--max-wait <seconds>no1800Hard ceiling for —watch in seconds (default 1800 = 30 min)
--include <kinds>no""Comma-separated extras to attach when —watch terminates: payload (full per-execution snapshot, can be hundreds of KB)
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow experiment|exp cancel [options] <workflow-id> <batchId>

Cancel every execution in a batch. Idempotent.

Arguments

NameRequiredVariadicDescription
workflow-idyesno
batchIdyesno

Options

FlagRequiredDefaultDescription
--yesnoRequired for non-TTY shells (CI, pipes). Acts immediately, no prompt.
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow experiment|exp results [options] <workflow-id> [batchId]

Download eval results in CSV or JSON.

Arguments

NameRequiredVariadicDescription
workflow-idyesno
batchIdnono

Options

FlagRequiredDefaultDescription
--format <csv|json>yesOutput format
--out <path>noOutput file. When omitted, the binary streams to stdout.
--base-url <url>noServer base URL

eigenpal workflow experiment|exp compare|diff [options] <batchIdA> <batchIdB>

Diff eval scores between two experiment batches.

Arguments

NameRequiredVariadicDescription
batchIdAyesno
batchIdByesno

Options

FlagRequiredDefaultDescription
--sort <abs-delta-desc|delta-asc|delta-desc|name>no"abs-delta-desc"Row sort order (default: biggest movers first)
--regression-threshold <n>no0.05Δ below this is flagged as a regression (default 0.05)
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow experiment|exp watch [options] <workflow-id> <batchId>

Poll until terminal, then auto-pull results, replaces status --watch + results --out.

Arguments

NameRequiredVariadicDescription
workflow-idyesno
batchIdyesno

Options

FlagRequiredDefaultDescription
--interval <seconds>no5Poll interval in seconds (default 5)
--max-wait <seconds>no1800Hard ceiling in seconds (default 1800 = 30 min)
--pull-on-complete <path>noDestination for the results file. Default: ./results-<batchId>.<format>
--format <csv|json>no"json"Results export format (default json)
--no-pullnoSkip auto-pulling results on terminal (just watch)
--base-url <url>noServer base URL

eigenpal workflow versions list|ls [options] <workflow-id>

List historical workflow versions, newest first.

Arguments

NameRequiredVariadicDescription
workflow-idyesno

Options

FlagRequiredDefaultDescription
--limit <n>no50Page size
--offset <n>no0Page offset
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow versions restore [options] <workflow-id> <versionId>

Restore the workflow to a previous version.

Arguments

NameRequiredVariadicDescription
workflow-idyesno
versionIdyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow step-type list|ls [options]

List every step type the deployment supports.

Options

FlagRequiredDefaultDescription
--search <q>noFilter
--limit <n>no50Page size
--offset <n>no0Page offset
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow step-type get [options] <type>

Return the full schema and behavioral docs for one step type.

Arguments

NameRequiredVariadicDescription
typeyesno

Options

FlagRequiredDefaultDescription
--out <path>noWrite to file

eigenpal workflow evaluator-type list|ls [options]

List every evaluator type with a one-line description.

Options

FlagRequiredDefaultDescription
--search <q>noFilter by type, name, or description
--limit <n>no50Page size
--offset <n>no0Page offset
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON

eigenpal workflow evaluator-type get [options] <type>

Fetch the JSON Schema for one evaluator type. Pipe through jq to inspect specific fields.

Arguments

NameRequiredVariadicDescription
typeyesno

Options

FlagRequiredDefaultDescription
--out <path>noWrite to file

eigenpal workflow step exec [options] <type>

DISABLED, local mimic runners removed pending server-side redesign (EIG-104). Use run or workflow experiment run instead.

Arguments

NameRequiredVariadicDescription
typeyesno

Options

FlagRequiredDefaultDescription
--config-json <json>no(unused; kept for back-compat parsing)
--config-file <path>no(unused)
--inputs <pairs...>no(unused)
--output-schema <path>no(unused)