Skip to main content

Commands

Core

CommandDescription
eigenpal runs list|ls [options] [source]List runs across workflows and agents, optionally scoped to one source.
eigenpal runs get [options] <run-id>Get one run.
eigenpal runs compare|diff [options] <reference-run-id> <run-id>Compare one run against another run. PDF/DOCX text comparison uses pdftotext/python3 and reports byte fallbacks.
eigenpal runs rerun [options] <run-id>Create a new run from a previous run’s stored input snapshot.
eigenpal runs trace [options] <run-id>Print raw trace.jsonl for a run, or write it with —out.
eigenpal runs watch [options] <run-id>Watch a run until it reaches a terminal status.
eigenpal runs cancel [options] <run-id>Cancel a run.

Artifacts

CommandDescription
eigenpal runs artifacts|artifact list|ls [options] <run-id>List available agent run artifacts without downloading them.
eigenpal runs artifacts|artifact fetch [options] <run-id>Download agent run artifacts by canonical artifact path.

Feedback

CommandDescription
eigenpal runs feedback|fb update [options] <run-id>Edit feedback state, rating, message, or expected JSON for a run.
eigenpal runs feedback|fb resolve [options] <run-id>Mark run feedback as resolved.
eigenpal runs feedback|fb clear [options] <run-id>Delete feedback, expected.json, and expected files for a run.

Expected

CommandDescription
eigenpal runs expected list|ls [options] <run-id>List expected JSON and files attached to a run.
eigenpal runs expected pull [options] <run-id>Download expected JSON and files attached to a run.
eigenpal runs expected upload [options] <run-id> <file>Upload a local file as an expected artifact.
eigenpal runs expected copy-output [options] <run-id> <output-file>Copy a generated output file into expected artifacts.
eigenpal runs expected rename [options] <run-id> <old-name> <new-name>Rename an expected artifact.
eigenpal runs expected delete [options] <run-id> <name>Delete an expected artifact.

Details

eigenpal runs list|ls [options] [source]

List runs across workflows and agents, optionally scoped to one source.

Arguments

NameRequiredVariadicDescription
sourcenono

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--limit <n>no50Page size
--offset <n>no0Page offset
--jsonnoOutput the raw server response as JSON
--type <type>noFilter by run type: workflow|agent
--status <status>noFilter by run status
--source-ref <ref>noFilter agent runs by source ref
--batch-id <id>noFilter eval/experiment runs by batch id
--example-id <id>noFilter eval runs by example id
--example-id-contains <text>noFilter eval runs by example id substring
--from <date>noCreated at or after this ISO timestamp or relative date
--to <date>noCreated before this ISO timestamp or relative date
--created-after <date>noAlias for —from
--created-before <date>noAlias for —to
--completed-after <date>noCompleted at or after this ISO timestamp or relative date
--completed-before <date>noCompleted before this ISO timestamp or relative date
--compactnoRender compact run rows
--sort <field>noSort field; only createdAt is supported
--order <asc|desc>noSort order; only desc is supported

eigenpal runs get [options] <run-id>

Get one run.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--step <name>noFor workflow runs, show only this step (or comma-separated list)
--include <parts>no"feedback"Comma-separated extra parts or workflow step fields: feedback,expected,files,trace,issues,input,output,error,duration,inputRef

eigenpal runs compare|diff [options] <reference-run-id> <run-id>

Compare one run against another run. PDF/DOCX text comparison uses pdftotext/python3 and reports byte fallbacks.

Arguments

NameRequiredVariadicDescription
reference-run-idyesno
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--baselinenoCompare actual outputs from both runs instead of expected artifacts
--step <name>noFor workflow runs, restrict comparison to one step
--out <dir>noWrite comparison artifacts to this directory
--normalize-datesnoNormalize YYYYMMDD and YYYY-MM-DD tokens in filenames/text
--fail-on-diffnoExit 1 when comparison status is fail

eigenpal runs rerun [options] <run-id>

Create a new run from a previous run’s stored input snapshot.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--source-ref <ref>noSource ref for the new run: latest (default) or original. Workflow runs support only latest/original.
--waitnoPoll until the rerun reaches a terminal status
--interval <seconds>no2Polling interval in seconds
--max-wait <seconds>no1800Maximum wait before exiting 2

eigenpal runs trace [options] <run-id>

Print raw trace.jsonl for a run, or write it with —out.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--out <file>noOutput file path

eigenpal runs watch [options] <run-id>

Watch a run until it reaches a terminal status.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--interval <seconds>no2Polling interval in seconds
--max-wait <seconds>no1800Maximum wait before exiting 2

eigenpal runs cancel [options] <run-id>

Cancel a run.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--yesnoRequired in non-interactive environments

eigenpal runs artifacts|artifact list|ls [options] <run-id>

List available agent run artifacts without downloading them.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

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

eigenpal runs artifacts|artifact fetch [options] <run-id>

Download agent run artifacts by canonical artifact path.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--out <dir>noOutput directory
--include <parts>no"all"Comma-separated parts: output,input,metadata,issues,trace,lockfile,expected,all
--path <path>no[]Fetch one exact artifact path from artifacts list; repeatable
--jsonnoOutput a JSON summary of written artifacts

eigenpal runs feedback|fb update [options] <run-id>

Edit feedback state, rating, message, or expected JSON for a run.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--status <open|resolved|ignored>noSet feedback status
--rating <pass|fail|partial|none>noSet feedback rating
--message <text>noSet feedback message body
--message-file <path>noRead feedback message body from a file
--expected-json <json>noSet structured expected JSON
--expected-json-file <path>noRead structured expected JSON from a file
--clear-messagenoClear the feedback message body
--clear-ratingnoClear feedback rating
--clear-expected-jsonnoDelete structured expected JSON

eigenpal runs feedback|fb resolve [options] <run-id>

Mark run feedback as resolved.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--message <text>noSet feedback message body
--message-file <path>noRead feedback message body from a file

eigenpal runs feedback|fb clear [options] <run-id>

Delete feedback, expected.json, and expected files for a run.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--yesnoRequired in non-interactive environments

eigenpal runs expected list|ls [options] <run-id>

List expected JSON and files attached to a run.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

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

eigenpal runs expected pull [options] <run-id>

Download expected JSON and files attached to a run.

Arguments

NameRequiredVariadicDescription
run-idyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--out <dir>noOutput directory

eigenpal runs expected upload [options] <run-id> <file>

Upload a local file as an expected artifact.

Arguments

NameRequiredVariadicDescription
run-idyesno
fileyesno

Options

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

eigenpal runs expected copy-output [options] <run-id> <output-file>

Copy a generated output file into expected artifacts.

Arguments

NameRequiredVariadicDescription
run-idyesno
output-fileyesno

Options

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

eigenpal runs expected rename [options] <run-id> <old-name> <new-name>

Rename an expected artifact.

Arguments

NameRequiredVariadicDescription
run-idyesno
old-nameyesno
new-nameyesno

Options

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

eigenpal runs expected delete [options] <run-id> <name>

Delete an expected artifact.

Arguments

NameRequiredVariadicDescription
run-idyesno
nameyesno

Options

FlagRequiredDefaultDescription
--base-url <url>noServer base URL
--jsonnoOutput the raw server response as JSON
--yesnoRequired in non-interactive environments