What causes "failed to resolve session: %w" in charmbracelet/crush?

List existing sessions (`crush sessions`) and copy the exact id

failed to resolve session: %w

Error message

failed to resolve session: %w

What it means

Wraps errors from resolveSession(), which finds the session for a non-interactive run based on --continue (session id) or --last flags. Thrown when no matching session exists or session lookup fails.

Source

Thrown at internal/cmd/run.go:237

		}
	}

	// Wait for the agent to become ready (MCP init, etc).
	if err := waitForAgent(ctx, c, ws.ID); err != nil {
		stopSpinner()
		return fmt.Errorf("agent not ready: %w", err)
	}

	// Force-update agent models so MCP tools are loaded.
	if err := c.UpdateAgent(ctx, ws.ID); err != nil {
		slog.Warn("Failed to update agent", "error", err)
	}

	defer stopSpinner()

	sess, err := resolveSession(ctx, c, ws.ID, continueSessionID, useLast)
	if err != nil {
		return fmt.Errorf("failed to resolve session: %w", err)
	}
	if continueSessionID != "" || useLast {
		slog.Info("Continuing session for non-interactive run", "session_id", sess.ID)
		// If no explicit model override was requested, restore the
		// model/provider from the last assistant message in the
		// session, provided it is still available.
		if largeModel == "" && smallModel == "" {
			if err := restoreModelFromSession(ctx, c, ws, sess.ID); err != nil {
				slog.Warn("Failed to restore model from session", "error", err)
			}
		}
	} else {
		slog.Info("Created session for non-interactive run", "session_id", sess.ID)
	}

	events, err := c.SubscribeEvents(ctx, ws.ID)
	if err != nil {
		return fmt.Errorf("failed to subscribe to events: %w", err)

View on GitHub (pinned to 7944b8e522)

Solutions

List existing sessions (`crush sessions`) and copy the exact id
Omit --continue/--last to create a new session instead
Run from the same project directory used when the session was created
Check the local SQLite database exists and is readable

Example fix

// before
crush run --continue abc123 "continue"
// after: use the exact session id
crush run --continue 0195f2a1-7c3b-7e22-9d4e-1a2b3c4d5e6f "continue"

Defensive patterns

Strategy: fallback

Validate before calling

// verify the session exists before asking for --continue
if continueSessionID != "" {
	if _, err := q.GetSession(ctx, continueSessionID); err != nil {
		return fmt.Errorf("session %q not found; omit --continue to start fresh", continueSessionID)
	}
}

Try / catch

sess, err := resolveSession(ctx, c, ws.ID, continueSessionID, useLast)
if err != nil {
	if useLast || continueSessionID != "" {
		slog.Warn("Session not found, starting a new one")
		sess, err = createNewSession(ctx, c, ws.ID)
	}
	if err != nil {
		return fmt.Errorf("failed to resolve session: %w", err)
	}
}

Prevention

Copy session ids from `crush sessions` output, not from memory
Run from the same project directory the session was created in
Fall back to a new session in scripted/CI usage

When it happens

Trigger: `crush run --continue <id>` / `--last` where resolveSession(ctx, c, ws.ID, continueSessionID, useLast) cannot find the session in the SQLite store or the query errors.

Common situations: Session id typo or session belongs to a different workspace/project; database moved or deleted (fresh clone); --last used when no prior sessions exist; wrong working directory so a different workspace ID is derived.

Related errors

AI-assisted analysis of charmbracelet/crush@7944b8e522 (2026-08-29). Data as JSON: /api/errors/be7424e258aaae5c. Report an issue: GitHub.