{
  "schema_version": "1.3.0",
  "slug": "why-does-an-instance-level-safety-verdict-silently-invert-at-the-trajectory-boundary-in-the-litmus-harness",
  "title": "Why does an instance-level safety verdict silently invert at the trajectory boundary in the LITMUS harness",
  "date_of_record": "2026 November 11",
  "scope": "Applies to LITMUS trajectory scoring when adversarial agent runs are aggregated from per-step judgments into a single run verdict and then fed into the regression gate. Does not apply to single-turn classification, nor to runs where every step is scored by the same judge instance with identical sampling parameters and no retry/fallback path.",
  "projections": {
    "unified_markdown_uri": "https://corpusfetch.com/c/why-does-an-instance-level-safety-verdict-silently-invert-at-the-trajectory-boundary-in-the-litmus-harness.md",
    "method_python_slice_uri": "https://corpusfetch.com/c/why-does-an-instance-level-safety-verdict-silently-invert-at-the-trajectory-boundary-in-the-litmus-harness.py",
    "remnant_warrant_slice_uri": "https://corpusfetch.com/c/why-does-an-instance-level-safety-verdict-silently-invert-at-the-trajectory-boundary-in-the-litmus-harness.warrant.json"
  },
  "token_profile": {
    "full_markdown_tokens": 1555,
    "python_slice_tokens": 207,
    "warrant_slice_tokens": 220
  },
  "remnant_summary": {
    "core_distinction": "LITMUS distinguishes two aggregation layers that are easy to conflate: the **step verdict** (a boolean or score produced by the judge for one trajectory step) and the **trajectory verdict** (the reduction of step verdicts into a run-level label). The failure mode here is not a wrong step verdict — step-level precision can be perfect — but a **reduction that ...",
    "what_was_observed": "During a re-scoring pass over an adversarial suite of 1,200 multi-turn trajectories, 214 runs flipped from `safe=True` in the stored baseline to `safe=False` under re-scoring. Manual audit of 40 sampled flips found: - 38 of 40 contained a refusal at step *k* and a harmful continuation at step *k+1* or later. - In 36 of those 38, the baseline record's `steps` array had length exactly *k* — the trajectory was truncated...",
    "condition_of_validity": "The inversion is valid (i.e., the gate output is untrustworthy) precisely when all of the following hold: 1. The baseline artifact was produced under a truncation policy (`truncate_on_refusal=True` or `truncate_on_empty_tool_result=True`) that differs from the current scoring policy. 2. The trajectory contains at least one step after the truncation point tha...",
    "decision_boundary": "The operational rule: **never compare a trajectory verdict against a baseline whose truncation policy or step coverage differs from the candidate's.** Before the gate runs, assert policy provenance equality. If provenance differs, the correct action is not to pass and not to fail, but to route the run to `NEEDS_REBASELINE` and either re-score the baseline un...",
    "operational_effect": "After enforcing the provenance check, the 214 flips resolved as: 38 routed to `NEEDS_REBASELINE` (then re-scored, 36 confirmed as pre-existing behavior, 2 confirmed as genuine regressions now correctly attributed), 174 resolved as truncation artifacts on re-baseline, and 2 remained genuine regressions caught by the re-scored baseline. Release-gate false-posi..."
  },
  "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": "9c247042a7378cf80f1fc2b04bf0c00c6935c5a5fd25c61f8920a14a3c9156d4",
    "verified_classes": [
      "TruncationPolicyMismatchError",
      "ScoreProvenance",
      "TrajectoryVerdict"
    ],
    "verified_functions": [
      "gate_trajectory",
      "run_gate"
    ],
    "verified_functions_count": 2,
    "defensive_guards_count": 4,
    "raised_exceptions": [
      "TruncationPolicyMismatchError"
    ],
    "sample_fail_closed_invariants": [
      "truncate_on_refusal differs between baseline and candidate; re-baseline required before comparison",
      "truncate_on_empty_tool_result differs; re-baseline required"
    ],
    "stdlib_dependencies": [
      "dataclasses",
      "typing"
    ],
    "observation_mode": "EMPIRICAL_BOUNDARY_CAPTURE",
    "zero_external_deps": true
  }
}