Update to Marketing Swarm

{
  "name": "sandbox-python-executor",
  "description": "Use when Marketing Swarm needs deterministic metric calculation, tabular analysis, simulation, statistical checks, file processing, or verification that should actually run with host-native Python instead of relying on unverified mental arithmetic.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 294
    }
  ],
  "skill_md_contents": "---\nname: sandbox-python-executor\ndescription: Use when Marketing Swarm needs deterministic metric calculation, tabular analysis, simulation, statistical checks, file processing, or verification that should actually run with host-native Python instead of relying on unverified mental arithmetic.\n---\n\n# Sandbox Python Executor\n\nUse the host's own Python execution capability to produce evidence, not just code suggestions.\n\nThis Skill does not create a remote runtime and does not declare an MCP dependency. Tool availability belongs to the host.\n\n## Use Python for\n\n- campaign metric calculation across many rows\n- period comparisons and decompositions\n- scenario and sensitivity analysis\n- bootstrap or Monte Carlo calculations when justified\n- incrementality/lift statistics\n- CSV/JSON parsing and data validation\n- deterministic charts/tables when requested\n- archive/package verification during Plugin maintenance\n\n## Execution rule\n\n1. Actually execute the calculation when Python is available and the answer depends on it.\n2. Keep source campaign files read-only unless the user requested transformation.\n3. State assumptions and data-cleaning choices that affect the result.\n4. Do not assume sandbox internet access.\n5. Do not expose tokens, credentials, or unrelated files.\n6. Preserve generated artifacts the user needs and return the host-provided file reference/path when available.\n\n## Evidence\n\nReport enough to distinguish an executed calculation from an estimate: operation, important inputs/filters, pass/fail or result summary, generated file when applicable, and warnings.\n\n## If Python is unavailable\n\nDo not claim execution occurred. Continue with static reasoning only when appropriate and mark execution-dependent calculations as unverified.\n"
}