multica-ai/multica · error

search skills: %w

Error message

search skills: %w

What it means

Returned by `multica skill search` when GET `/api/skills/search?q=<query>` fails. The query is URL-escaped client-side, so encoding is not the issue; the wrapped cause is transport failure, timeout (60s budget covers the server's search/index work), auth rejection, or a server-side search error (e.g. index not built).

Source

Thrown at server/cmd/multica/cmd_skill.go:612

func runSkillSearch(cmd *cobra.Command, args []string) error {
	client, err := newAPIClient(cmd)
	if err != nil {
		return err
	}

	query := strings.TrimSpace(args[0])
	if query == "" {
		return fmt.Errorf("query is required")
	}

	ctx, cancel := context.WithTimeout(context.Background(), cli.AtLeastAPITimeout(60*time.Second))
	defer cancel()

	var results []map[string]any
	path := "/api/skills/search?q=" + url.QueryEscape(query)
	if err := client.GetJSON(ctx, path, &results); err != nil {
		return fmt.Errorf("search skills: %w", err)
	}

	output, _ := cmd.Flags().GetString("output")
	if output == "json" {
		return cli.PrintJSON(os.Stdout, results)
	}

	headers := []string{"NAME", "URL", "SOURCE", "INSTALLS", "DESCRIPTION"}
	rows := make([][]string, 0, len(results))
	for _, result := range results {
		rows = append(rows, []string{
			strVal(result, "name"),
			strVal(result, "url"),
			strVal(result, "source"),
			strVal(result, "install_count"),
			strVal(result, "description"),
		})
	}

View on GitHub (pinned to 2c0912b6ec)

Solutions

1. Verify basic API health with `multica skill list` — if that works, the failure is specific to search.
2. Retry once; index warm-up and transient timeouts often clear immediately.
3. Use a more specific query to reduce server-side search cost.
4. If list works but every search 500s, check server logs for index-related errors.

Example fix

# before
multica skill search a   # times out over huge catalog

# after
multica skill list > /dev/null && multica skill search "ascii tables"

multica skill list > /dev/null || exit 1   # pre-flight API reachability
[ "${#QUERY}" -ge 2 ] || { echo 'use a query of 2+ chars'; exit 1; }

Retry once on timeout-shaped errors (cold index, transient load); do not retry 401/403 (fix auth) or persistent 500s (inspect server logs for the search index). Wrap the CLI call in a script, capture stderr, and branch on status keywords before deciding to retry.

When it happens

Trigger: Server down or wrong --api-url; auth token expired (401); search index cold or corrupted causing 500; extremely broad query (single common character) overloading the server until the context deadline expires.

Common situations: First search after server startup before the index warms; CLI configured for a stale environment; large skill catalog with a one-character query; transient network blip between CLI and server.

Related errors

• update skill: %w
• delete skill: %w
• refresh skill: %w
• import skill: %w
• query is required

AI-assisted analysis of multica-ai/multica@2c0912b6ec (2026-08-15). Data as JSON: /api/errors/e407445643a2d0a1. Report an issue: GitHub.