Detects the host and harnesses, installs/updates a pinned automation toolset, hardens every tool and harness against telemetry, and applies safe allow/ask/deny permission profiles. Use for setting up or upgrading the AI dev environment, hardening tool privacy, or configuring harness permissions.
Ship orchestrator — classify the project, pick a mode (improve/release/launch/maintain), delegate only the skills the project's signals justify, consolidate findings, produce an audit report. Use for an end-to-end audit or ship pipeline across multiple skills.
Approved simplification — overengineering hygiene with one approval batch and zero silent deletion. Use when removing complexity, dead code, or overengineering safely.
Autonomous optimization — iterate on any project with a measurable metric (Karpathy autoresearch pattern). Use when optimizing toward a metric via automated experiment loops.
Plan executor — take an issue, a specs/{feature}/tasks.md, or a plain request and implement it unit by unit with a verify signal per unit, red-proven tests, budgeted backtracking, and a code-proven close. Use when the plan exists and the work is to be done, not planned.
Bug hunter — reproduce the failure as an observed red, localize it (bisect, logs, tracing), test at most three hypotheses, land the minimal fix behind a red-proven regression test, and prove it with the project's own check. Use when something is broken and the cause is not yet known.
Release cutter — version from the commits, dated CHANGELOG section, every version surface bumped, project check green, local release commit + annotated tag; every publishing step (push, GitHub release, registry, store) handed to the human with the exact command. Use when a release is to be cut.
★not rated 1 4d agoA64 tokens
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: