@eigenpal/cli 0.10.6 → 0.10.7
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/CHANGELOG.md +6 -0
- package/dist/cli.js +2673 -305
- package/dist/skill/SKILL.md +16 -3
- package/package.json +1 -1
package/dist/skill/SKILL.md
CHANGED
|
@@ -20,7 +20,10 @@ appear in user examples, but do not prefer them.
|
|
|
20
20
|
experiments, versions, and step schemas.
|
|
21
21
|
- `eigenpal agents` manages Git-backed agent source packages, shared resources,
|
|
22
22
|
releases, sync, secrets, datasets, and live file inspection.
|
|
23
|
-
- `eigenpal run` starts a workflow or agent run from a target.
|
|
23
|
+
- `eigenpal run` starts a single workflow or agent run from a target. This is the
|
|
24
|
+
command for running something once and reading its output, including
|
|
25
|
+
API-triggered workflows. (`eigenpal workflow experiment` is a separate batch run
|
|
26
|
+
over a whole dataset, not the single-run path.)
|
|
24
27
|
- `eigenpal runs` inspects, watches, compares, reruns, cancels, and downloads
|
|
25
28
|
artifacts for workflow or agent runs.
|
|
26
29
|
|
|
@@ -48,10 +51,20 @@ eigenpal workflow step-type get ai.extract --json | jq '.configSchema.properties
|
|
|
48
51
|
eigenpal workflow validate ./workflow.yaml
|
|
49
52
|
eigenpal workflow push --file workflow.yaml
|
|
50
53
|
|
|
51
|
-
# Run
|
|
54
|
+
# Run with ad-hoc input and read the output (use this for API-triggered workflows).
|
|
55
|
+
# `--wait --json` polls to completion and prints the run with top-level `output`.
|
|
56
|
+
eigenpal run workflows.<workflow-id> --input-json '{"document":"..."}' --wait --json | jq '.output'
|
|
57
|
+
eigenpal run workflows.<workflow-id> --input-file document=./invoice.pdf --wait --json | jq '.output'
|
|
58
|
+
|
|
59
|
+
# Run one persisted example instead.
|
|
52
60
|
eigenpal run workflows.<workflow-id> --example <example-name>
|
|
61
|
+
|
|
62
|
+
# Inspect a run after the fact. `runs get --json` returns top-level
|
|
63
|
+
# `output`/`files`/`error` once the run is completed. Plain `run --wait`
|
|
64
|
+
# without `--json` only prints "Run <id> is completed", so pass `--json` or
|
|
65
|
+
# use `runs get` to see the output.
|
|
53
66
|
eigenpal runs watch <execution-id>
|
|
54
|
-
eigenpal runs get <execution-id> --json
|
|
67
|
+
eigenpal runs get <execution-id> --json | jq '.output'
|
|
55
68
|
|
|
56
69
|
# Push dataset and evaluators.
|
|
57
70
|
eigenpal workflow dataset validate ./dataset
|