rseng-maintenance-sustainability

# Maintaining and sustaining research software Use this skill when the goal is to keep software usable over time rather than to ship a first version: setting up maintenance habits, managing dependencies and technical debt, deciding whether to keep, deprecate, or archive a project, and keeping its sustainability story honest. Unmaintained software degrades even with no code changes - dependencies age, environments shift, and the knowledge to run it erodes - so treat maintenance as a recurring cost, not a one-off. ## Establish maintenance habits early Calibrate effort to the user base, but build the habits before the software is widely used: - Write a test suite and check coverage with a language-appropriate tool (pytest-cov for Python, covr for R). Without tests, every dependency or environment update carries unknown regression risk. - Set up a CI pipeline that runs tests on every commit and on a schedule (e.g. weekly). Scheduled runs catch breakage from external changes even when nobody is actively developing. See the rseng-ci-cd and rseng-testing skills for the mechanics. - Keep documentation current as part of maintenance, not after it. If you cannot install and run the software from scratch using only its README and install guide, the docs need updating. See rseng-documentation. - Maintain a visible issue tracker (GitHub/GitLab Issues) so the maintenance backlog is shared, not held in one person's head. ## Manage dependencies deliberately Every depend...