evolve skill

Upload a skill folder, then attach it to a job. ./my-skill stands for your folder containing SKILL.md:

evolve skill upload ./my-skill

Use the skill name returned by the upload in --skill. For a skill named my-skill:

evolve run -d deepswe@1.1 \
  -i anko-default-function-arguments \
  -a codex -m gpt-6-luna \
  --skill name:my-skill \
  --max-trial-spend 1 --max-retries 0 \
  --watch

This singular skill group manages uploaded content. To read the CLI’s bundled manual, use evolve skills.

Upload

evolve skill upload ./my-skill --org acme

--org <name> selects the owning organization. Without it, the CLI uses its saved default, then your personal organization. A directory containing multiple skills can return multiple upload records. Each record has two ways to reference it:

Reference	Use when…
`upload:<id>`	You want this exact upload.
`name:<skill-name>`	You want the current upload under that name when the job is created.

Organization members can reference uploaded skills in their jobs. Only the creator can delete an upload. With --json, upload returns one skill record, or an array when multiple skills are uploaded. Each record includes id, name, and ref.

List and inspect

Set $SKILL_ID to the ID returned by upload or list:

evolve skill list --scope org
evolve skill show name:my-skill
evolve skill show "$SKILL_ID" --json

show includes metadata and SKILL.md. With --json, it returns one skill object; list returns a page. skill list accepts --scope <my|shared|org> and the shared list options.

Delete

evolve skill delete "$SKILL_ID"

A skill referenced by an unfinished job is refused with skill_in_use; wait for the job to finish or cancel it first. Past jobs keep their resolved skill records. This command has no confirmation prompt. With --json, successful deletion returns id and deleted: true. See exit codes for failures. Global options apply. See Skills for supported sources and how jobs resolve them.

Start work

Inspect results

Manage resources

Upload

List and inspect

Delete