Menú
Build multi-profile stress tests from brute complexity and constraints, then execute with traceable evidence.
Define and enforce project-wide quality standards for agent and skill documents. Use when creating, reviewing, or refactoring files under agents/ and skills/ to keep structure, terminology, and output contracts consistent.
Validate statement samples and sample files for competitive programming problems. Ensures the expected outputs in problem statements match the actual solution output.
Audit std/brute assumptions with MCP evidence, including worst/average complexity risk and stress readiness.
Audit statement, tutorial, and samples for consistency and publication readiness before packaging.
Verify final tests with hard quality gates: integrity, consistency, validator, limit semantics, and wrong-solution kill.
Use before coding to decide whether a problem idea is judgeable, implementable, and verifiable.
| name | stress-strategy |
| description | Build multi-profile stress tests from brute complexity and constraints, then execute with traceable evidence. |
| disable-model-invocation | false |
Goal: build a reproducible and explainable multi-profile stress strategy from brute complexity and constraints, instead of running a single random profile.
tiny_exhaustive, random_small, and edge_small.type=1 for tiny_exhaustive, type=2 for random_small, and type=3/4 for edge_small.type=3: boundary and extreme constraints coverage.type=4: targeted worst-case/TLE-inducing patterns, not simple max-parameter scaling.O(n^2), automatically lower n_max and trials.When calling stress_test_run, pass stress_profiles and record completed rounds and failure points for each profile.
decision: go / no_gostress_profiles: each profile's name/trials/types/generator_argsexecution_summary: completion status and failed rounds per profilenext_fix_hint: if failed, provide prioritized fix direction (std/brute/generator/validator)go: all required profiles complete and no correctness mismatch remains.no_go: any required profile is incomplete, or mismatches remain unresolved.type=4 as a duplicate of type=3.