adversarial-search-campaign

name	adversarial-search-campaign
description	Analyze adversarial route/search campaigns with canonical commands, expected artifacts, status boundaries, and claim limits.
category	campaign-analysis
kind	analysis
phase	analysis
requires_write	true
requires_slurm	false
requires_benchmark_artifacts	true
delegates_to	["benchmark-row-status","artifact-provenance","evidence-synthesis"]
output_schema	benchmark_analysis_summary.v1

Adversarial Search Campaign

When to use

Use this skill for the campaign lane named by the title when you need canonical command surfaces, expected artifacts, row-status handling, and conservative claim boundaries.

Command surfaces

Start with scripts/tools/compare_adversarial_samplers.py, scripts/tools/generate_adversarial_routes.py, and adversarial failure archives. Prefer config-first commands and existing validation launch-packet scripts when available.

Workflow

1. Record route sampler, scenario split, seed schedule, failure archive, and fallback/degraded counts.
2. Capture commit SHA, config path, command, output root, seeds, and expected artifacts.
3. Classify each planner/scenario row with benchmark-row-status when benchmark rows exist.
4. Classify generated outputs with artifact-provenance.
5. Feed multi-source conclusions through evidence-synthesis.

Guardrails

• Do not count fallback or degraded runs as success evidence.
• Do not make paper-facing or population-level claims beyond the inspected evidence.
• Keep raw generated outputs out of git unless they are deliberate compact fixtures.

Output

Use the output_schema declared in frontmatter.