akarachen/long-running-agent
Skills for long-running agent work — write the completion contract, carry a goal to a verified finish, keep review adversarial, and orchestrate large migrations.
Runs independent adversarial review and correction loops to catch failures beyond compilation and passing tests. Use when an agent produces a large volume of consequential code, configs, or migrations that will be merged or released.
Organizes large migrations around shared decisions, mechanical translation, and ordered executable gates. Use when a language, framework, or architecture migration is too large to perform file by file by hand.
Use when a goal is already approved and must be carried to a verified finish — freeze an outcome plan, judge completion from evidence, and stop when the same gap repeats. Use when an agent would otherwise declare the goal done, rewrite the objective, raise the bar between rounds, or keep going after the same failure.
Carries a large agent effort from a countable completion contract through production, independent review, and verified finish. Use when one agent run or supervised effort must complete a long refactor, migration, backlog, bulk generation, or cleanup task.
Help the user craft a well-specified /goal objective for goal mode — turn a rough intention into a completion contract with a clear finish line, proof, boundaries, and stop rule. Use when the user asks for help writing, refining, or improving a goal.