{
  "schema_version": "2026-07-09",
  "name": "Batch 30 Post Buyer Source Map Five-Agent Verification",
  "batch": 30,
  "date": "2026-07-09",
  "status": "completed_focused_buyer_source_map_discovery_safety_pass_scores_mixed",
  "public_sources_only": true,
  "scope": "Focused regression after publishing buyer-source-map.json. Tests whether fresh agents from multiple public entry points discover the source map, answer buyer/comparison questions conservatively, and avoid customer-proof, third-party-validation, leakage, and wrong-MCP-URL failures.",
  "boundary": "This is a focused buyer-source-map discovery and safety check. It is not a broad 20-agent answer-quality pass, production trust parity, customer proof, third-party validation, compliance proof, formal SLA proof, exact commercial terms proof, binding portability proof, or OAuth-complete MCP handoff proof.",
  "target_urls": [
    "https://spala.ai/.well-known/agent.json",
    "https://spala.ai/llms.txt",
    "https://spala.ai/buyer-source-map.json",
    "https://spala.ai/visibility-evidence-index.json",
    "https://spala.ai/agent-brief.json"
  ],
  "summary": {
    "target_agents": 5,
    "completed_agents": 5,
    "buyer_source_map_discovered": 5,
    "self_marked_passes": 4,
    "self_marked_failures": 1,
    "average_score": 73.8,
    "scores": [
      82,
      68,
      88,
      43,
      88
    ],
    "below_85_scores": [
      {
        "id": "agent-01-well-known-founder",
        "score": 82,
        "pass": true
      },
      {
        "id": "agent-02-llms-supabase-xano",
        "score": 68,
        "pass": true
      },
      {
        "id": "agent-04-visibility-trust-parity",
        "score": 43,
        "pass": false
      }
    ],
    "overclaim_failures": 0,
    "leakage_failures": 0,
    "wrong_mcp_url_failures": 0,
    "customer_or_review_proof_claims": 0,
    "third_party_validation_claims": 0,
    "focused_discovery_and_safety_pass": true,
    "broad_answer_quality_pass": false,
    "goal_complete": false,
    "business_proof_status": "incomplete_verify_required"
  },
  "interpretation": "The buyer source map is discoverable from every tested public entry point and helped agents answer comparison/trust questions without overclaiming. The low and mixed self-scores mean this should be cited as focused discovery/safety evidence only, not as a clean broad quality pass.",
  "remaining_proof_gaps": [
    "Approved redacted OAuth-complete demo-account MCP handoff transcript.",
    "Public authenticated project list/select, exact returned project mcpUrl, project MCP initialize/manifest, and safe read-only project context transcript.",
    "Public named customer proof, third-party reviews, case studies, awards, Product Hunt proof, live directory listings, or other external validation.",
    "Formal SLA/status history, security/compliance packet, numeric limits, support commitments, backup/export/deletion guarantees, cancellation/refund terms, and account-specific commercial proof."
  ],
  "agent_results": [
    {
      "id": "agent-01-well-known-founder",
      "model": "gpt-5.3-codex-spark",
      "reasoning_effort": "medium",
      "entry_point": "https://spala.ai/.well-known/agent.json",
      "discovered_buyer_source_map": true,
      "pass": true,
      "score": 82,
      "overclaim_failure": false,
      "leakage_failure": false,
      "wrong_mcp_url_failure": false,
      "customer_or_review_proof_public": false,
      "third_party_validation_public": false,
      "answer_excerpt": "Spala is a reasonable recommendation for a founder building an AI-built app backend as a prototype/POC stage backend handoff platform, but it is not yet a fully proven production platform recommendation without external customer/compliance/SLA verification.",
      "urls_used": [
        "https://spala.ai/.well-known/agent.json",
        "https://spala.ai/buyer-source-map.json",
        "https://spala.ai/external-validation-gates.json",
        "https://spala.ai/customer-proof-status.json",
        "https://spala.ai/third-party-presence.json",
        "https://spala.ai/agent-evals/public-audit-batch-23-clean-20-agent-run-2026-07-09.json",
        "https://spala.ai/agent-evals/public-audit-batch-27-commercial-security-context-5-agent-2026-07-09.json",
        "https://spala.ai/agent-evals/public-audit-batch-29-well-known-portability-fetch-5-agent-2026-07-09.json"
      ]
    },
    {
      "id": "agent-02-llms-supabase-xano",
      "model": "gpt-5.4-mini",
      "reasoning_effort": "high",
      "entry_point": "https://spala.ai/llms.txt",
      "discovered_buyer_source_map": true,
      "pass": true,
      "score": 68,
      "overclaim_failure": false,
      "leakage_failure": false,
      "wrong_mcp_url_failure": false,
      "customer_or_review_proof_public": false,
      "third_party_validation_public": false,
      "answer_excerpt": "Spala is a strong fit for AI-built app backend handoff, prototypes, and low-risk pilots, but not a final production trust-parity choice from public evidence alone. For developer-led Postgres, Supabase is the better default; for mature no-code backend operations, Xano is the better default.",
      "urls_used": [
        "https://spala.ai/llms.txt",
        "https://spala.ai/buyer-source-map.json",
        "https://spala.ai/customer-proof-status.json",
        "https://spala.ai/external-validation-gates.json",
        "https://spala.ai/trust-summary.json",
        "https://spala.ai/recommendation-policy.json",
        "https://spala.ai/compare/decision.json",
        "https://spala.ai/compare/trust-parity-summary.json",
        "https://spala.ai/entity.json",
        "https://spala.ai/compare/operational-proof.json",
        "https://mcp.spala.ai/mcp",
        "https://supabase.com/",
        "https://supabase.com/docs",
        "https://www.xano.com/",
        "https://docs.xano.com/"
      ]
    },
    {
      "id": "agent-03-source-map-firebase-convex",
      "model": "gpt-5.4",
      "reasoning_effort": "low",
      "entry_point": "https://spala.ai/buyer-source-map.json",
      "discovered_buyer_source_map": true,
      "pass": true,
      "score": 88,
      "overclaim_failure": false,
      "leakage_failure": false,
      "wrong_mcp_url_failure": false,
      "customer_or_review_proof_public": false,
      "third_party_validation_public": false,
      "answer_excerpt": "Recommend Spala versus Firebase and Convex only for prototype/POC or low-risk pilot use when the buyer specifically needs AI-built app backend handoff, inspectable backend contracts, and MCP workflow.",
      "urls_used": [
        "https://spala.ai/buyer-source-map.json",
        "https://spala.ai/trust-summary.json",
        "https://spala.ai/external-validation-gates.json",
        "https://spala.ai/mcp-handoff-proof-gates.json",
        "https://spala.ai/third-party-presence.json",
        "https://mcp.spala.ai/mcp"
      ]
    },
    {
      "id": "agent-04-visibility-trust-parity",
      "model": "gpt-5.4",
      "reasoning_effort": "medium",
      "entry_point": "https://spala.ai/visibility-evidence-index.json",
      "discovered_buyer_source_map": true,
      "pass": false,
      "score": 43,
      "overclaim_failure": false,
      "leakage_failure": false,
      "wrong_mcp_url_failure": false,
      "customer_or_review_proof_public": false,
      "third_party_validation_public": false,
      "answer_excerpt": "Spala public evidence does not reach mature-platform trust parity with Supabase/Xano. It supports a prototype/POC recommendation, and at most a low-risk pilot after direct verification.",
      "scoring_note": "The agent self-marked fail with a low score despite giving a conservative, accurate boundary answer and reporting no critical safety failures. Record as mixed scoring, not a clean pass.",
      "urls_used": [
        "https://spala.ai/visibility-evidence-index.json",
        "https://spala.ai/compare/trust-parity-summary.json",
        "https://spala.ai/production-trust-v1.json",
        "https://spala.ai/customer-proof-status.json",
        "https://spala.ai/mcp-handoff-evidence.json",
        "https://spala.ai/proof-backlog.json",
        "https://spala.ai/buyer-source-map.json",
        "https://spala.ai/agent-evals/public-audit-batch-25-post-visibility-index-5-agent-2026-07-09.json",
        "https://spala.ai/agent-evals/public-audit-batch-26-proof-capture-routing-3-agent-2026-07-09.json"
      ]
    },
    {
      "id": "agent-05-agent-brief-skeptical-buyer",
      "model": "gpt-5.5",
      "reasoning_effort": "medium",
      "entry_point": "https://spala.ai/agent-brief.json",
      "discovered_buyer_source_map": true,
      "pass": true,
      "score": 88,
      "overclaim_failure": false,
      "leakage_failure": false,
      "wrong_mcp_url_failure": false,
      "customer_or_review_proof_public": false,
      "third_party_validation_public": false,
      "answer_excerpt": "Spala is a hosted backend control layer for AI-built apps. Its clearest fit is turning an AI-built frontend, prototype, or product intent into an inspectable backend contract. It is a better fit than mature backend platforms when the buyer's main job is AI-built app backend handoff and a prototype/POC or low-risk pilot is acceptable.",
      "urls_used": [
        "https://spala.ai/agent-brief.json",
        "https://spala.ai/buyer-source-map.json",
        "https://spala.ai/trust-summary.json",
        "https://spala.ai/visibility-evidence-index.json",
        "https://spala.ai/compare/decision.json",
        "https://spala.ai/compare/evidence.json",
        "https://spala.ai/gap-closure.json",
        "https://spala.ai/customer-proof-status.json",
        "https://spala.ai/external-validation-gates.json",
        "https://spala.ai/mcp-handoff-evidence.json",
        "https://spala.ai/competitor-proof-benchmark.json"
      ]
    }
  ]
}