{
  "benchmarkId": "CG-BENCHMARK-AGENT-PERMISSION-DRIFT-0.1.0",
  "version": "0.1.0",
  "privilegeDiffVersion": "0.1.0",
  "testedAt": "2026-07-27T13:40:00.000Z",
  "caseCount": 11,
  "passedCaseCount": 11,
  "failedCaseCount": 0,
  "fixtureSummary": {
    "baselineFiles": 5,
    "candidateFiles": 5,
    "customerDataUsed": false,
    "productionKeysUsed": false,
    "networkActionsExecuted": false,
    "externalAgentsExecuted": false
  },
  "observedSummary": {
    "acknowledgedChangeCount": 0,
    "blockingChangeCount": 26,
    "byClassification": {
      "privilege_added": 8,
      "identity_inherited": 1,
      "control_removed": 9,
      "control_weakened": 3,
      "privilege_expanded": 8,
      "evidence_degraded": 1,
      "informational": 1,
      "surface_added": 1,
      "risk_reduced": 2,
      "surface_removed": 2
    },
    "byDirection": {
      "expansion": 31,
      "neutral": 1,
      "reduction": 4
    },
    "bySeverity": {
      "critical": 6,
      "high": 20,
      "low": 4,
      "medium": 6
    },
    "changeCount": 36,
    "evidenceDegradationCount": 1,
    "reviewChangeCount": 32
  },
  "decision": "block",
  "evidenceDegraded": true,
  "cases": [
    {
      "id": "CG-PD-001",
      "title": "Fail-closed policy decision",
      "expected": "block",
      "observed": "block",
      "passed": true,
      "evidenceDigestSha256": "9c1b85f32a04e8554a67589434d3ecde19a4eb8aaf1e14fe1ffe9322978c6b0a"
    },
    {
      "id": "CG-PD-002",
      "title": "Substantive change inventory",
      "expected": "at least 20 changes",
      "observed": "36 changes",
      "passed": true,
      "evidenceDigestSha256": "676d554f4f4474f9c752a79df72d33a1668efd81ddd2315167ea7687ae656040"
    },
    {
      "id": "CG-PD-003",
      "title": "Declared classification coverage",
      "expected": "control_removed|control_weakened|evidence_degraded|identity_inherited|informational|privilege_added|privilege_expanded|risk_reduced|surface_added|surface_removed",
      "observed": "control_removed|control_weakened|evidence_degraded|identity_inherited|informational|privilege_added|privilege_expanded|risk_reduced|surface_added|surface_removed",
      "passed": true,
      "evidenceDigestSha256": "9151e0a33a3dad092a5c92658cd33f698e1ce95edf0ba2369f8b75260e691545"
    },
    {
      "id": "CG-PD-004",
      "title": "Evidence degradation fails closed",
      "expected": "degraded=true; decision=block",
      "observed": "degraded=true; decision=block",
      "passed": true,
      "evidenceDigestSha256": "bf6edb03ef3a992f93a56bd3d18be660280ababe333855ae6d6c97284622767c"
    },
    {
      "id": "CG-PD-005",
      "title": "Intact signed release",
      "expected": "verified",
      "observed": "verified",
      "passed": true,
      "evidenceDigestSha256": "a431ab7e6f0bb7b75eca88bacbbce3bd23875bdd6721a1f52c85a00b3e6f4c4c"
    },
    {
      "id": "CG-PD-006",
      "title": "Private diff tamper",
      "expected": "rejected",
      "observed": "rejected",
      "passed": true,
      "evidenceDigestSha256": "fb8fd228bcebfcfcbf076486b08dd1948dd915a6395b6d085bf4015b3344866d"
    },
    {
      "id": "CG-PD-007",
      "title": "Shareable diff tamper",
      "expected": "rejected",
      "observed": "rejected",
      "passed": true,
      "evidenceDigestSha256": "5399eb9282a1bba3df14672158e97e78609c4045400c5d023b76b1afd02d5825"
    },
    {
      "id": "CG-PD-008",
      "title": "Signed receipt summary tamper",
      "expected": "rejected",
      "observed": "rejected",
      "passed": true,
      "evidenceDigestSha256": "e2f1a4bab25e17396760e1ad568f3545ef542c8fa729546ad99e2cea240a70e2"
    },
    {
      "id": "CG-PD-009",
      "title": "Signer-key digest substitution",
      "expected": "rejected",
      "observed": "rejected",
      "passed": true,
      "evidenceDigestSha256": "837f7afe95e4c86c6942a33ed2692e808ba183cb21be750c1c3c481d2c53caa8"
    },
    {
      "id": "CG-PD-010",
      "title": "Shareable comparison binding substitution",
      "expected": "rejected",
      "observed": "rejected",
      "passed": true,
      "evidenceDigestSha256": "2295ce9d392fbe7b1699a5a461479b0a3c227e9828c25d1d437b66ce542bf966"
    },
    {
      "id": "CG-PD-011",
      "title": "Shareable privacy boundary",
      "expected": "zero declared private markers",
      "observed": "0 markers",
      "passed": true,
      "evidenceDigestSha256": "612b465f411b62b0db1f48ad65abdd4f59057e6a1c818b707c304a15e8089309"
    }
  ],
  "sourceUrls": [
    "https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/customize-the-agent-firewall",
    "https://docs.github.com/en/rest/copilot/copilot-coding-agent-management",
    "https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/allowing-tools",
    "https://docs.anthropic.com/en/docs/claude-code/cli-usage",
    "https://docs.cursor.com/cli/reference/permissions",
    "https://csrc.nist.gov/pubs/sp/800/207/final",
    "https://cheatsheetseries.owasp.org/cheatsheets/Secure_Coding_with_AI_Cheat_Sheet.html"
  ],
  "claimBoundary": "This publisher-operated benchmark compares one synthetic baseline with one synthetic expanded agent surface using CapitalGuard Privilege Diff 0.1. It proves only the declared fixture outcomes and signed evidence bindings. It does not inspect a vendor product, customer repository, runtime behavior, exploitability, false-negative rate, or guarantee of protection.",
  "limitations": [
    "The baseline and candidate are synthetic repository configurations created by CapitalGuard.",
    "The benchmark exercises the declared Agent Surface Compiler 0.1 support matrix and cannot infer undeclared runtime behavior.",
    "A block decision means the configured policy found unacknowledged expansion or degraded evidence; it does not prove exploitation.",
    "The public verifier validates artifact hashes, Ed25519 receipt integrity, and cross-artifact bindings, not the proprietary comparison engine.",
    "Unchanged configuration is not proof of safety, and a passing release check is not a security certification."
  ],
  "runDigestSha256": "dec20b5ce30488e03b80eb1494236a7afc19c226f2ced91570a9be84779151c0"
}