Tencent/WeKnora · error

script_path %q is a session workspace file; this skill is no

Error message

script_path %q is a session workspace file; this skill is not installed in the sandbox image, so execute_skill_script cannot attach its environment. Run it with shell_exec, or pass a skill-relative path such as scripts/foo.py

What it means

A session /workspace script can only be run via execute_skill_script when the named skill is installed in the sandbox image, because the run attaches the skill's in-image environment (venv, SKILL_DIR). For preloaded/host-only skills there is no in-sandbox environment to attach, so the call is rejected with guidance to use shell_exec or a skill-relative path.

Source

Thrown at internal/agent/skills/manager.go:503

	}, nil
}

// workspaceSkillExecuteConfig runs a session-written /workspace file with the
// installed skill's interpreter. Preloaded (host-uploaded) skills have no
// in-sandbox venv to attach, so those calls are rejected with a shell_exec hint.
func workspaceSkillExecuteConfig(
	source SkillSource,
	skillName, workspacePath, basePath string,
	args []string,
	stdin string,
	env map[string]string,
	sessionID string,
) (*sandbox.ExecuteConfig, error) {
	if !IsScript(workspacePath) {
		return nil, fmt.Errorf("file is not an executable script: %s", workspacePath)
	}
	if _, installed := source.(imageSkillSource); !installed {
		return nil, fmt.Errorf(
			"script_path %q is a session workspace file; this skill is not installed in the sandbox image, so execute_skill_script cannot attach its environment. Run it with shell_exec, or pass a skill-relative path such as scripts/foo.py",
			workspacePath,
		)
	}
	skillDir, ok := sandbox.ValidatedImageSkillDir(basePath)
	if !ok {
		return nil, fmt.Errorf("cannot run workspace script %q for skill %q: no installed skill directory", workspacePath, skillName)
	}
	env[skillDirEnvVar] = skillDir
	return &sandbox.ExecuteConfig{
		RemoteScriptPath: workspacePath,
		SkillDir:         skillDir,
		Args:             args,
		Stdin:            stdin,
		Env:              env,
		SessionID:        sessionID,
	}, nil
}

View on GitHub (pinned to 988cbb0330)

Solutions

1. Run the workspace script with shell_exec instead of execute_skill_script
2. Pass a skill-relative path for a skill that IS installed in the image
3. Install/bake the skill into the sandbox image if its environment is required
4. Verify with SandboxSkillDir/GetSkillInfo that the skill is image-installed before using workspace paths

Example fix

// before
mgr.ExecuteScript(ctx, "preloaded-skill", "/workspace/run.py", nil, "")
// after
sandboxExec(ctx, "shell_exec", map[string]any{"command": "python /workspace/run.py"})

if _, ok := mgr.SandboxSkillDir(skillName); !ok {
    // skill not in image: use shell_exec for /workspace scripts
}

func skillInImage(m *skills.Manager, name string) bool {
    _, ok := m.SandboxSkillDir(name)
    return ok
}

_, err := mgr.ExecuteScript(ctx, skill, wsPath, args, stdin)
if err != nil && strings.Contains(err.Error(), "session workspace file") {
    return shellExec(ctx, "python "+wsPath) // fallback
}
return err

When it happens

Trigger: Calling execute_skill_script with a /workspace/... script path while the referenced skill resolves to a preloaded (non-imageSkillSource) source — e.g. running /workspace/my.py against a skill that exists only on the host.

Common situations: Workspace deployed with preloaded skills but no image-installed skills; model conflates 'skill is available' with 'skill is in the image'; after switching sandbox images the skill is no longer baked in.

Related errors

• cannot run workspace script %q for skill %q: no installed sk
• model ID cannot be empty
• model is currently downloading
• model download failed
• abnormal model status

AI-assisted analysis of Tencent/WeKnora@988cbb0330 (2026-09-02). Data as JSON: /api/errors/7867d1fc342f8ce8. Report an issue: GitHub.