GoogleContainerTools/skaffold · error

couldn't find profile %s

Error message

couldn't find profile %s

What it means

A profile name was activated but no profile with that name exists in the parsed skaffold configuration. ApplyProfiles builds a name->profile map and, when an activated profile (via CLI `-p`, auto-activation, or dependency chains) is missing from the map, it throws `couldn't find profile <name>`.

Source

Thrown at pkg/skaffold/schema/profiles.go:54

	skutil "github.com/GoogleContainerTools/skaffold/v2/pkg/skaffold/util"
	"github.com/GoogleContainerTools/skaffold/v2/pkg/skaffold/util/stringslice"
	"github.com/GoogleContainerTools/skaffold/v2/pkg/skaffold/yaml"
	"github.com/GoogleContainerTools/skaffold/v2/pkg/skaffold/yamltags"
)

// ApplyProfiles modifies the input skaffold configuration by the application
// of a list of profiles, and returns the list of applied profiles.
func ApplyProfiles(c *latest.SkaffoldConfig, fieldsOverrodeByProfile map[string]configlocations.YAMLOverrideInfo, opts cfg.SkaffoldOptions, namedProfiles []string) ([]string, map[string]configlocations.YAMLOverrideInfo, error) {
	byName := profilesByName(c.Profiles)

	profiles, contextSpecificProfiles, err := activatedProfiles(c.Profiles, opts, namedProfiles)
	if err != nil {
		return nil, nil, fmt.Errorf("finding auto-activated profiles: %w", err)
	}
	for _, name := range profiles {
		profile, present := byName[name]
		if !present {
			return nil, nil, fmt.Errorf("couldn't find profile %s", name)
		}

		if err := applyProfile(c, fieldsOverrodeByProfile, profile); err != nil {
			return nil, nil, fmt.Errorf("applying profile %q: %w", name, err)
		}
	}

	// remove profiles section for run modes where profiles are already merged into the main pipeline
	switch opts.Mode() {
	case cfg.RunModes.Build, cfg.RunModes.Dev, cfg.RunModes.Deploy, cfg.RunModes.Debug, cfg.RunModes.Render, cfg.RunModes.Run, cfg.RunModes.Diagnose, cfg.RunModes.Delete:
		c.Profiles = nil
	}
	return profiles, fieldsOverrodeByProfile, checkKubeContextConsistency(contextSpecificProfiles, opts.KubeContext, c.Deploy.KubeContext)
}

func checkKubeContextConsistency(contextSpecificProfiles []string, cliContext, effectiveContext string) error {
	// cli flag takes precedence
	if cliContext != "" {

View on GitHub (pinned to a1189de023)

Solutions

1. Check the exact profile name defined under `profiles:` in skaffold.yaml and correct the `-p` value
2. Run `skaffold config` / inspect skaffold.yaml to list available profile names
3. If the profile lives in another config fragment, use `--profile` with the right module context or include the fragment
4. Fix `requires: profiles` references in other profiles that point at the missing name

Example fix

# before
skaffold dev -p prodcta
# after
skaffold dev -p prod

func requireProfiles(cfg *latest.SkaffoldConfig, wanted []string) error {
    have := map[string]bool{}
    for _, p := range cfg.Profiles { have[p.Name] = true }
    for _, w := range wanted {
        if !have[strings.TrimPrefix(w, "-")] {
            return fmt.Errorf("profile %q not defined in skaffold.yaml", w)
        }
    }
    return nil
}

When it happens

Trigger: Running `skaffold dev -p myprofile` (or render/build/deploy) where `myprofile` is not defined under `profiles:` in skaffold.yaml; a profile dependency (`requires: profiles`) referencing a nonexistent name; a typo in a `-p` flag value or in a cross-file profile reference.

Common situations: Typo in the `-p` flag; profile defined in a skaffold.yaml fragment not included in the current config; renaming a profile without updating dependents/CI scripts; multi-module setups where the profile exists only in another module.

Related errors

• getting run context: %w
• finding auto-activated profiles: %w
• applying profile %q: %w
• value must be one of `always`, `missing`, or `never`
• CONFIG_REMOTE_REPO_CACHE_NOT_FOUND_ERR

AI-assisted analysis of GoogleContainerTools/skaffold@a1189de023 (2026-09-05). Data as JSON: /api/errors/dfac38ab0eefdbbf. Report an issue: GitHub.