What problem does it solve?
Long-running autonomous agents and parallel multi-agent orchestrations fail when their launch prompts are vague: agents return answer-shaped near misses, converge prematurely on one approach, fabricate progress, or give up early. This Skill provides a rigorous brief-writing methodology so expensive agent runs are steered by precise success predicates, enumerated non-counting outcomes, and adversarial verification gates.
Core Features & Use Cases
- Pseudo-Formal Task Briefs: Structure launch prompts with definitions covering degenerate cases, an exact success predicate with quantifiers, enumerated non-counting outcomes, and domain-specific auditor checklists.
- Parallel Orchestration Policy: Manage diverse portfolios of parallel workers with an approach-family registry, blocked-route bookkeeping, early independence, and late cross-pollination.
- Persistence and Return Gates: Pair effort floors and solvability framing with verification gates, and phrase return conditions as predicates over the artifact rather than agent confidence.
- Use Case: Before launching a 64-agent orchestration on an open research problem, use this Skill to convert a vague goal into a brief that pre-blocks the likely near misses, equips adversarial auditors with a failure-mode checklist, and gates the return on surviving audit.
Quick Start
Ask the agent to convert your hard problem statement into a pseudo-formal long-horizon task brief with a success predicate, non-counting outcomes, and an adversarial audit checklist before launching the run.
