mithril-studio
- 5 skills
- 0 followers
- 4 hours ago last updated
- ▌ Memory · mithril-studioRead accumulated project learnings at session start and record new ones before finishing. Use at the beginning of every task to load what past sessions learned about this repo, and at the end to write down conventions, failures and their fixes, patterns, and decisions worth carrying forward. Triggers on "what do we know about", "record this learning", "prime memory", or any task in a repo containing a .mem/ directory.
- ▌ Factory Compose · mithril-studio bundleTurn a project brief into an ordered backlog of dependency-sequenced GitHub issues for the Software Factory, each labeled agent:queued so the factory builds them lowest-number-first. Drafts the full ordered issue list from a brief plus a target owner/repo, shows it for human review and editing, and only creates issues after explicit approval. Use when the user says "compose a backlog", "factory-compose", "plan issues for the factory", "decompose this brief into issues", or hands over a project brief/spec together with a target owner/repo.
- ▌ Bounded Waits · mithril-studioCap how long you passively babysit a backgrounded long-running command (a full e2e suite, a build, a slow integration run) before moving on. Use whenever you background a command and start polling for it to finish, especially near the end of a session — never let watching a slow check consume the time or turn budget you need to commit and open the PR.
- ▌ Checkpoint Commits · mithril-studioCommit and push every working increment as you go, instead of holding all work uncommitted until the session ends. Use during any multi-step implementation session — especially long or unattended ones (a factory run, a background agent, anything that can be killed by a timeout, a crash, or a lost connection) — where uncommitted work has nowhere else to live.
- ▌ Performance Optimization · mithril-studioOptimizes application performance across frontend, backend, queries, and databases. Use when performance requirements exist, when you suspect performance regressions, when Core Web Vitals or load times need improvement, when N+1 query patterns need fixing, or when profiling reveals bottlenecks.