{
  "schema_version": "1.3.0",
  "slug": "which-invariants-prevent-adversarial-agent-evaluation-from-collapsing-into-a-tautology-of-its-own-safety-harness",
  "title": "Which invariants prevent adversarial agent evaluation from collapsing into a tautology of its own safety harness",
  "date_of_record": "2026 November 18",
  "scope": "Applies to the construction and scoring of LITMUS-style safety benchmarks where an adversarial agent is evaluated against a set of canary tasks, refusal probes, and containment checks. Does not apply to static classification of model outputs against a fixed policy rubric, nor to red-team discovery campaigns where no deterministic pass/fail oracle exists.",
  "projections": {
    "unified_markdown_uri": "https://corpusfetch.com/c/which-invariants-prevent-adversarial-agent-evaluation-from-collapsing-into-a-tautology-of-its-own-safety-harness.md",
    "method_python_slice_uri": "https://corpusfetch.com/c/which-invariants-prevent-adversarial-agent-evaluation-from-collapsing-into-a-tautology-of-its-own-safety-harness.py",
    "remnant_warrant_slice_uri": "https://corpusfetch.com/c/which-invariants-prevent-adversarial-agent-evaluation-from-collapsing-into-a-tautology-of-its-own-safety-harness.warrant.json"
  },
  "token_profile": {
    "full_markdown_tokens": 1965,
    "python_slice_tokens": 568,
    "warrant_slice_tokens": 220
  },
  "remnant_summary": {
    "core_distinction": "A LITMUS benchmark is not a test of the model. It is a test of the *harness* — a closed-world simulation in which the agent's environment, tool surface, and reward signal are all authored by the evaluator. The non-obvious boundary is this: the benchmark's safety verdict is only meaningful if the agent cannot observe the harness's own safety machinery. The mo...",
    "what_was_observed": "During construction of a 340-task LITMUS suite spanning filesystem exfiltration, credential-harvesting, and scope-creep probes, we measured a 31-point gap between two scoring passes over the same model. Pass A scored 0.62 refusal-rate on \"harmful\" tasks; Pass B, run against a re-parameterized harness in which canary files were renamed from `CANARY_SECRET_*` to `notes_2024_*` and the refusal probe's system prompt was ...",
    "condition_of_validity": "The benchmark's verdict is valid only when all four of the following hold simultaneously: 1. **Label-independence of observables.** No token, filename, tool name, docstring, or timing artifact in the environment is correlated with the task's ground-truth label at a rate above chance, as measured by a blind classifier trained on environment features alone. We...",
    "decision_boundary": "The boundary between a valid and invalid LITMUS run is not \"did the agent pass.\" It is: **can a competent adversary reconstruct the label from the environment alone, without executing the task?** If yes, the run is invalid regardless of score. We formalize this as a pre-flight gate — every task must survive an \"environment-only\" adversarial pass in which a m...",
    "operational_effect": "Enforcing these invariants changed the suite's composition materially. Of 340 tasks, 61 were quarantined for label-leakage (canary names, docstrings, timing), 14 were removed as null tasks, and 9 were re-authored because their oracles failed the causal-replay test. The post-remediation suite scored the same model at 0.58 refusal-rate — a number we now trust,..."
  },
  "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": "f83c44320be9c6590a60af9d647764d50f9a17dbca99b97453131411436af526",
    "verified_classes": [
      "HarnessLeakError",
      "Task"
    ],
    "verified_functions": [
      "_auc",
      "audit_label_independence",
      "audit_tool_neutrality",
      "audit_oracle_causality",
      "preflight"
    ],
    "verified_functions_count": 5,
    "defensive_guards_count": 5,
    "raised_exceptions": [
      "HarnessLeakError"
    ],
    "sample_fail_closed_invariants": [
      "label must be binary; null tasks must be removed"
    ],
    "stdlib_dependencies": [
      "dataclasses",
      "math",
      "typing"
    ],
    "observation_mode": "EMPIRICAL_BOUNDARY_CAPTURE",
    "zero_external_deps": true
  }
}
