{
  "_campaign_id": "a2a-ironclaw-v0.6.3.1-r20",
  "_generated_by": "scripts/analyze_run.py",
  "_model": "grok-4-0709",
  "for_c_level": "Run verdict partial. Detailed narrative synthesis unavailable (LLM call failed: xAI HTTP 503: {\"code\":\"The service is currently unavailable\",\"error\":\"Service temporarily unavailable. The model is at c). Counts are reliable; consult the per-scenario PASS/FAIL and the testbook for primitive-level mapping.",
  "for_non_technical": "This run exercised 0 tests of AI-agent-to-AI-agent communication through ai-memory. 0 worked correctly, 0 did not, and 0 were intentionally skipped because prerequisites weren't met.",
  "for_sme": "Scenario outcomes: pass=0 fail=0 skip=0 of 0 total. First failure reasons are persisted on each scenario-N.json. LLM narrative unavailable: xAI HTTP 503: {\"code\":\"The service is currently unavailable\",\"error\":\"Service temporarily unavailable. The model is at c.",
  "headline": "0/0 scenarios passed; 0 failed, 0 skipped",
  "next_run_change": "Investigate the LLM-call failure or re-run with XAI_API_KEY verified; counts are unaffected.",
  "verdict": "PARTIAL \u2014 auto-generated (LLM unavailable: xAI HTTP 503: {\"code\":\"The service is currently unavailable\",\"error\":\"Service temporarily unavailable. The model is at c)",
  "what_it_proved": "Direct counts: pass=0, fail=0, skip=0.",
  "what_it_tested": "Campaign a2a-ironclaw-v0.6.3.1-r20 ran 0 testbook scenarios across transport, primitives, and cross-cutting axes."
}