Self-Admitted Technical Debt (SATD) is a particular case of Technical Debt (TD) in which developers rely on source code comments (SATD-C) or labeled issues (SATD-I) to report their sub-optimal technical solutions. In this paper, we first explore a sample of 286 SATD-I instances collected from five open source projects, including Microsoft Visual Studio Code and GitLab Community Edition. We show that in 45% of the studied issues TD was introduced to ship earlier (i.e., to deliver faster), and in almost 60% it refers to Design flaws. Besides, we report that most developers pay SATD-I to reduce its costs or interests (66%). To complement the previous exploratory results, we investigate the adoption of tools to support SATD-I documentation. For that, we build a large-scale dataset of 72K SATD-C and 20K SATD-I instances, extracted from 190 well-known GitHub projects. We also implement a prototype tool, called AdmiTD, to automatically report SATD-C as GitHub issues. We use this dataset and tool to reveal that developers are not interested in the automatic transformation of SATD-C in SATD-I. Moreover, we show that it might not be feasible to create a tool to recommend explicit links between SATD-C and SATD-I instances.