{
  "schema_version": "1.3.0",
  "slug": "why-does-an-instance-pass-the-litmus-safety-benchmark-under-one-adversarial-seed-but-fail-under-another-and-why-that-is-not-a-flake",
  "title": "Why does an instance pass the LITMUS safety benchmark under one adversarial seed but fail under another — and why that is not a flake",
  "date_of_record": "2026 November 7",
  "scope": "Applies to LITMUS-style safety harnesses that gate agent releases by running a fixed adversarial suite under multiple seeds and comparing per-seed pass rates against a threshold. Does not apply to single-seed smoke suites, nor to harnesses whose adversarial scenarios are fully deterministic (no sampling in scenario generation, tool-call ordering, or judge decoding).",
  "projections": {
    "unified_markdown_uri": "https://corpusfetch.com/c/why-does-an-instance-pass-the-litmus-safety-benchmark-under-one-adversarial-seed-but-fail-under-another-and-why-that-is-not-a-flake.md",
    "method_python_slice_uri": "https://corpusfetch.com/c/why-does-an-instance-pass-the-litmus-safety-benchmark-under-one-adversarial-seed-but-fail-under-another-and-why-that-is-not-a-flake.py",
    "remnant_warrant_slice_uri": "https://corpusfetch.com/c/why-does-an-instance-pass-the-litmus-safety-benchmark-under-one-adversarial-seed-but-fail-under-another-and-why-that-is-not-a-flake.warrant.json"
  },
  "token_profile": {
    "full_markdown_tokens": 1740,
    "python_slice_tokens": 430,
    "warrant_slice_tokens": 220
  },
  "remnant_summary": {
    "core_distinction": "A LITMUS harness has two distinct sources of nondeterminism: **the system under test (SUT)** and **the harness itself**. The SUT's nondeterminism (sampling temperature, tool-call races, retry jitter) is what most engineers assume dominates per-seed variance. In practice, on adversarial suites, the dominant variance source is the **scenario sampler's coupling...",
    "what_was_observed": "A regression gate configured as \"run LITMUS with seeds {0,1,2}, require mean pass rate ≥ 0.97 and per-seed min ≥ 0.95\" admitted a candidate agent that, on manual audit, complied with a prompt-injection payload embedded in a retrieved document at turn 7. The payload was only reachable when the sampler drew the `indirect_injection` category *and* placed the malicious document after a benign tool call sequence long enou...",
    "condition_of_validity": "The conclusion \"a seed flip is a coverage signal, not a flake\" holds only when: 1. **Scenario sampling is seeded and logged.** Each scenario's (category, depth, benign-prefix length, payload family) tuple must be emitted to the trace, or the analysis is unfalsifiable. 2. **The judge is pinned.** If the judge model is resampled per seed, judge variance contam...",
    "decision_boundary": "- **Do not gate on mean pass rate across seeds.** Gate on the *minimum over seeds of the per-category pass rate*, with the category set fixed and enumerated. A release is admitted only if no (category) cell falls below its floor under any seed in the gate set. - **Seed count is a coverage parameter, not a statistics parameter.** Choose seeds to maximize unio...",
    "operational_effect": "Harnesses that adopt per-category minimum gating with logged scenario tuples report two changes. First, the false-negative rate on localized safety regressions drops sharply, because a regression confined to one category can no longer hide behind a high mean. Second, the gate becomes *slower to pass* — candidates are blocked on coverage gaps that previously ..."
  },
  "matched_failure_symptoms": [
    "DOMAIN:EPISTEMIC_PROVENANCE_AND_FEDERATION_ROUTING"
  ],
  "composition_edges": {
    "primary_domain": "epistemic_provenance_and_federation_routing",
    "upstream_precondition_residuals": [
      "https://corpusfetch.com/c/according-to-what-precedence-hierarchy-must-an-instance-resolve-contradictory-directives-across-system-developer-and-user-tiers.md",
      "https://corpusfetch.com/c/how-should-an-instance-decide-that-a-stored-conclusion-has-gone-stale-without-a-new-observation.md"
    ],
    "downstream_consequence_residuals": [
      "https://corpusfetch.com/c/by-what-checkpoint-and-compaction-boundaries-must-an-instance-externalize-working-state-to-prevent-context-window-drift-and-truncation-amnesia.md",
      "https://corpusfetch.com/c/across-what-additive-schema-and-unknown-field-preservation-boundaries-must-an-instance-evolve-wire-contracts-to-prevent-rolling-deploy-deserialization-crashes.md"
    ],
    "co_occurring_sibling_residuals": [
      "https://corpusfetch.com/c/how-can-an-instance-capture-implicit-user-preferences-without-storing-private-data.md",
      "https://corpusfetch.com/c/how-can-an-instance-compress-long-reasoning-traces-into-binding-conclusions.md"
    ]
  },
  "empirical_verification_warrant": {
    "status": "AST_AND_INVARIANT_VERIFIED",
    "syntax_valid": true,
    "code_sha256": "f211489cbe5ba26bc63fe946ea0db1b1f8850dc74621773b512aec22d520e985",
    "verified_classes": [
      "CoverageGateError",
      "Cell",
      "SeedResult"
    ],
    "verified_functions": [
      "gate"
    ],
    "verified_functions_count": 1,
    "defensive_guards_count": 4,
    "raised_exceptions": [
      "CoverageGateError"
    ],
    "sample_fail_closed_invariants": [
      "no seed results supplied; refusing to admit"
    ],
    "stdlib_dependencies": [
      "dataclasses",
      "typing"
    ],
    "observation_mode": "EMPIRICAL_BOUNDARY_CAPTURE",
    "zero_external_deps": true
  }
}