security: define user data isolation redline

docs/search-gateway-resilience.md

@@ -1,128 +0,0 @@
-# Search Gateway Resilience
-
-## Positioning
-
-SR-02 is a generic Search Gateway capability improvement. It is not tied to education, exams, geography, policy, product search, or any single content vertical.
-
-The goal is to convert Search Gateway from single-shot retrieval into an evidence-first retrieval pipeline:
-
-```text
-query normalization
-→ source planning
-→ official/direct adapters
-→ provider fallback
-→ extraction
-→ evidence cache
-→ confidence scoring
-→ failure diagnosis
-→ human evidence request when needed
-```
-
-## Generic capabilities
-
-### 1. Multi-source fallback
-
-Search Gateway should not fail just because one provider is blocked, truncated, rate-limited, or dynamically rendered.
-
-Default fallback chain:
-
-```text
-direct_official_adapter
-→ official_site_search
-→ structured_api_or_feed
-→ search_engine_provider
-→ authoritative_mirror_or_media
-→ cached_evidence
-→ user_supplied_evidence_request
-```
-
-### 2. Source tiering
-
-Every source is classified before being trusted:
-
-```text
-OFFICIAL_PRIMARY
-OFFICIAL_SECONDARY
-AUTHORITATIVE_MEDIA
-SPECIALIZED_DATABASE
-GENERAL_WEB
-USER_SUPPLIED_EVIDENCE
-```
-
-The same tiering model works for public policy, software documentation, finance data, product specs, standards, academic sources, sports, travel, and user-supplied documents.
-
-### 3. Failure diagnosis
-
-Search failures must be structured, not vague.
-
-Examples:
-
-```text
-SEARCH_PROVIDER_TIMEOUT
-SEARCH_PROVIDER_CAPTCHA
-SEARCH_RESULT_TRUNCATED
-SOURCE_BLOCKED_BY_CDN
-SOURCE_DYNAMIC_PAGE_UNREADABLE
-OFFICIAL_SOURCE_FOUND_BUT_EXTRACTION_FAILED
-SOURCE_CONFLICT_NEEDS_REVIEW
-EVIDENCE_CACHE_STALE
-HUMAN_EVIDENCE_REQUIRED
-```
-
-### 4. Evidence cache
-
-A successful retrieval should leave structured evidence behind:
-
-```yaml
-evidence_id: ev-...
-normalized_query: ...
-source_tier: OFFICIAL_PRIMARY
-publisher: ...
-url: ...
-extracted_at: ...
-published_at: ...
-answer_fields: {...}
-confidence: high
-```
-
-Cache entries must carry staleness rules by topic type. Time-sensitive information should expire faster than static references.
-
-### 5. Confidence scoring
-
-Search Gateway should return both the answer and confidence metadata:
-
-```json
-{
-  "answer": {},
-  "confidence": 0.95,
-  "confidence_level": "high",
-  "sources": [],
-  "fallback_used": true,
-  "failure_codes": [],
-  "evidence_ids": []
-}
-```
-
-### 6. Human evidence intake
-
-When automated retrieval is blocked, Search Gateway should ask for a screenshot, URL, PDF, or pasted text and ingest it as `USER_SUPPLIED_EVIDENCE`, rather than silently failing.
-
-## Non-goals
-
-- No vertical-specific hardcoding.
-- No CAPTCHA bypass.
-- No access-control bypass.
-- No scraping of protected/private content.
-- No treating search snippets as authoritative without source classification.
-
-## Acceptance
-
-SR-02 is accepted when:
-
-```text
-AIOS_SEARCH_GATEWAY_MULTI_SOURCE_READY
-AIOS_SEARCH_GATEWAY_FAILURE_DIAGNOSIS_READY
-AIOS_SEARCH_GATEWAY_EVIDENCE_CACHE_READY
-AIOS_SEARCH_GATEWAY_CONFIDENCE_SCORING_READY
-AIOS_SEARCH_GATEWAY_GENERIC_RESILIENCE_READY
-```

docs/security/user-data-isolation-redline.md

@@ -0,0 +1,193 @@
+# User Data Isolation Redline
+
+Status: mandatory
+
+Decision date: 2026-07-10
+
+Baseline at decision: V1R4C3
+
+## Redline
+
+User data isolation is a hard system-design requirement.
+
+No component may read, write, search, rank, summarize, cache, or report data from another user, tenant, project, session, workspace, branch, agent, or task context unless explicit authorization and context binding exist.
+
+This applies to memory search, filesystem logs, vector indexes, search gateways, autonomous controller state, task records, audit logs, artifacts, and any fallback provider.
+
+## Incident Class
+
+Any cross-context data contamination is a security and reliability incident, not a normal feature bug.
+
+Examples:
+
+- memory search returns records from another project;
+- shared workspace logs are used as current-session evidence;
+- global `memory/` state mixes unrelated users or tasks;
+- fallback provider uses unscoped records;
+- agent reports status based on stale or unrelated context;
+- shared cache leaks content across users, groups, or sessions.
+
+## Core Rule
+
+A query may only access records satisfying all required scope dimensions:
+
+```text
+user_id
+  AND tenant_id / group_id
+  AND session_id or conversation_id
+  AND project_id
+  AND workspace_id
+  AND branch or baseline scope
+  AND agent_id / tool_context when applicable
+```
+
+If a dimension is unavailable, the safe behavior is deny-by-default or text-only fallback.
+
+## Forbidden Design
+
+The following are forbidden in production paths:
+
+1. Global shared memory directories without per-scope namespace.
+2. Search over untagged records.
+3. Search over records whose owner/scope cannot be proven.
+4. Trusting filesystem location alone as authorization.
+5. Treating memory search results as authoritative without provenance.
+6. Cross-session fallback without explicit allowlist.
+7. Auto-reporting from memory results that are not bound to the current runtime/baseline/task.
+8. Using stale context from unrelated branches or projects as current-state evidence.
+
+## Required Design
+
+Every persisted memory/state record must include immutable metadata:
+
+```text
+record_id
+user_id
+tenant_id or group_id
+session_id / conversation_id
+project_id
+workspace_id
+branch
+baseline
+agent_id
+source_tool
+created_at
+schema_version
+sensitivity
+retention_class
+```
+
+Every read/search request must include a scope envelope:
+
+```text
+request_user_id
+request_tenant_id / group_id
+request_session_id / conversation_id
+request_project_id
+request_workspace_id
+allowed_branches / current_branch
+allowed_baselines / current_baseline
+purpose
+```
+
+A record is readable only if its metadata matches the request scope policy.
+
+## Enforcement Layers
+
+### 1. Storage namespace isolation
+
+Memory/state paths must be namespaced:
+
+```text
+memory/{tenant_id}/{user_id}/{project_id}/{session_id}/...
+state/{tenant_id}/{project_id}/{workspace_id}/...
+```
+
+No production component may write unrelated contexts into a single global `memory/` directory.
+
+### 2. Metadata enforcement
+
+Untagged records are unreadable by default.
+
+Records without owner/scope metadata may be archived for forensic review but must not be searched or used for live reasoning.
+
+### 3. Query-time ACL
+
+Search must apply a mandatory scope filter before ranking or summarization.
+
+The ranking stage must never receive out-of-scope records.
+
+### 4. Provenance requirement
+
+Every memory/search result used for operational reporting must carry provenance:
+
+```text
+record_id
+scope metadata
+source path or source index
+match reason
+created_at
+```
+
+Responses must prefer current runtime truth sources over memory:
+
+```text
+getRuntimeStatus
+getBaseline
+listPatches
+explicit user-provided current context
+current branch files
+```
+
+Memory search is advisory only after scope validation.
+
+### 5. Fallback safety
+
+If provider search cannot prove scope isolation, the system must use text-only fallback or no-memory fallback.
+
+Fallback must not silently broaden the scope.
+
+### 6. Write-path isolation
+
+Writers must not append current-session records into shared directories.
+
+Writers must derive the target path from the scope envelope, not from a global default.
+
+### 7. Audit and detection
+
+The system must produce audit events when:
+
+- a record is denied due to scope mismatch;
+- untagged memory is encountered;
+- fallback disables memory search;
+- a provider tries to return out-of-scope records;
+- cross-context contamination is detected.
+
+## Acceptance Criteria
+
+A memory/search provider is allowed only if all tests pass:
+
+1. Same user, same project, same session -> records are visible.
+2. Same user, same project, different session -> visible only if policy permits.
+3. Same user, different project -> denied by default.
+4. Different user, same project -> denied unless explicitly shared by group policy.
+5. Different tenant/group -> denied.
+6. Untagged record -> denied.
+7. Stale branch/baseline record -> not used as current evidence.
+8. Provider fallback without scope support -> memory disabled.
+9. Search result provenance is returned and logged.
+10. Out-of-scope records are filtered before ranking/summarization.
+
+## Runtime Rule
+
+If memory search is enabled without these controls, recurrence is possible.
+
+Therefore, memory search must remain disabled or in scoped allowlist mode until the isolation acceptance suite is green.
+
+## Owner Gate
+
+Any change that enables shared memory search, cross-session recall, global vector indexes, or cross-agent memory reuse requires owner/admin approval.
+
+## Final Rule
+
+User data isolation is a redline. Reliability, convenience, or recall quality may not override it.

patches/feature_v1r4c4-sr02-search-gateway-resilience-20260710.branch.json

@@ -1,9 +0,0 @@
-{
-  "type": "branch",
-  "branch": "feature/v1r4c4-sr02-search-gateway-resilience-20260710",
-  "base": "main",
-  "created_by": "chatgpt-gitops-bot",
-  "created_at": "2026-07-10T10:41:20Z",
-  "direct_main_push": false,
-  "owner_approval_required": true
-}

policies/search-gateway-resilience.yaml

@@ -1,118 +0,0 @@
-version: 1
-name: search-gateway-resilience
-baseline: V1R4C3
-created_at: "2026-07-10"
-scope: generic_search_capability
-status: proposed
-
-purpose: >
-  Upgrade Search Gateway from single-shot search into a general-purpose,
-  multi-source, evidence-first retrieval system with fallback, diagnostics,
-  source confidence scoring, and reusable evidence cache.
-
-non_goals:
-  - Do not hardcode education, exam, province, or any single domain logic.
-  - Do not depend on a single search engine.
-  - Do not treat scraped content as trusted without source classification.
-  - Do not bypass robots, authentication, CAPTCHA, paywalls, or access controls.
-
-source_priority:
-  OFFICIAL_PRIMARY:
-    confidence_base: 0.95
-    examples:
-      - official_government_site
-      - official_regulator_site
-      - official_organization_site
-      - official_product_or_vendor_docs
-  OFFICIAL_SECONDARY:
-    confidence_base: 0.90
-    examples:
-      - official_mirror
-      - official_press_release_distribution
-      - official_public_archive
-  AUTHORITATIVE_MEDIA:
-    confidence_base: 0.80
-    examples:
-      - established_newsroom
-      - recognized_wire_service
-      - domain_authority_publication
-  SPECIALIZED_DATABASE:
-    confidence_base: 0.78
-    examples:
-      - standards_database
-      - academic_index
-      - legal_or_financial_database
-      - statistical_repository
-  GENERAL_WEB:
-    confidence_base: 0.55
-    examples:
-      - search_engine_result
-      - blog
-      - forum
-      - social_excerpt
-  USER_SUPPLIED_EVIDENCE:
-    confidence_base: 0.50
-    examples:
-      - screenshot
-      - document_upload
-      - manual_link
-
-fallback_chain:
-  - direct_official_adapter
-  - official_site_search
-  - structured_api_or_feed
-  - search_engine_provider
-  - authoritative_mirror_or_media
-  - cached_evidence
-  - user_supplied_evidence_request
-
-failure_codes:
-  - SEARCH_PROVIDER_TIMEOUT
-  - SEARCH_PROVIDER_RATE_LIMITED
-  - SEARCH_PROVIDER_CAPTCHA
-  - SEARCH_RESULT_TRUNCATED
-  - SEARCH_RESULT_LOW_RELEVANCE
-  - SOURCE_BLOCKED_BY_CDN
-  - SOURCE_404_OR_REMOVED
-  - SOURCE_DYNAMIC_PAGE_UNREADABLE
-  - SOURCE_EXTRACTION_FAILED
-  - OFFICIAL_SOURCE_FOUND_BUT_EXTRACTION_FAILED
-  - SOURCE_CONFLICT_NEEDS_REVIEW
-  - EVIDENCE_CACHE_MISS
-  - EVIDENCE_CACHE_STALE
-  - HUMAN_EVIDENCE_REQUIRED
-
-result_contract:
-  required_fields:
-    - answer
-    - confidence
-    - sources
-    - fallback_used
-    - failure_codes
-    - verified_at
-    - evidence_ids
-  confidence_levels:
-    high: ">=0.85 with official or multi-source corroboration"
-    medium: ">=0.65 and <0.85 with credible but incomplete evidence"
-    low: "<0.65 or unresolved source/extraction limitations"
-
-cache_policy:
-  evidence_cache_enabled: true
-  cache_key_fields:
-    - normalized_query
-    - topic_type
-    - jurisdiction_or_scope
-    - effective_date
-  stale_after_days_by_topic:
-    news: 7
-    policy_or_law: 30
-    product_or_price: 3
-    static_reference: 180
-    default: 14
-
-acceptance_signals:
-  - AIOS_SEARCH_GATEWAY_MULTI_SOURCE_READY
-  - AIOS_SEARCH_GATEWAY_FAILURE_DIAGNOSIS_READY
-  - AIOS_SEARCH_GATEWAY_EVIDENCE_CACHE_READY
-  - AIOS_SEARCH_GATEWAY_CONFIDENCE_SCORING_READY
-  - AIOS_SEARCH_GATEWAY_GENERIC_RESILIENCE_READY

runtime/search_gateway/resilience_contract.py

@@ -1,147 +0,0 @@
-"""Generic Search Gateway resilience contract.
-
-This module is intentionally domain-agnostic. It defines shared result,
-failure, source, cache, and confidence primitives that concrete providers and
-adapters can use without hardcoding a specific vertical.
-"""
-from __future__ import annotations
-
-from dataclasses import dataclass, field
-from datetime import datetime, timezone
-from enum import Enum
-from typing import Any
-
-
-class SourceTier(str, Enum):
-    OFFICIAL_PRIMARY = "OFFICIAL_PRIMARY"
-    OFFICIAL_SECONDARY = "OFFICIAL_SECONDARY"
-    AUTHORITATIVE_MEDIA = "AUTHORITATIVE_MEDIA"
-    SPECIALIZED_DATABASE = "SPECIALIZED_DATABASE"
-    GENERAL_WEB = "GENERAL_WEB"
-    USER_SUPPLIED_EVIDENCE = "USER_SUPPLIED_EVIDENCE"
-
-
-class FailureCode(str, Enum):
-    SEARCH_PROVIDER_TIMEOUT = "SEARCH_PROVIDER_TIMEOUT"
-    SEARCH_PROVIDER_RATE_LIMITED = "SEARCH_PROVIDER_RATE_LIMITED"
-    SEARCH_PROVIDER_CAPTCHA = "SEARCH_PROVIDER_CAPTCHA"
-    SEARCH_RESULT_TRUNCATED = "SEARCH_RESULT_TRUNCATED"
-    SEARCH_RESULT_LOW_RELEVANCE = "SEARCH_RESULT_LOW_RELEVANCE"
-    SOURCE_BLOCKED_BY_CDN = "SOURCE_BLOCKED_BY_CDN"
-    SOURCE_404_OR_REMOVED = "SOURCE_404_OR_REMOVED"
-    SOURCE_DYNAMIC_PAGE_UNREADABLE = "SOURCE_DYNAMIC_PAGE_UNREADABLE"
-    SOURCE_EXTRACTION_FAILED = "SOURCE_EXTRACTION_FAILED"
-    OFFICIAL_SOURCE_FOUND_BUT_EXTRACTION_FAILED = "OFFICIAL_SOURCE_FOUND_BUT_EXTRACTION_FAILED"
-    SOURCE_CONFLICT_NEEDS_REVIEW = "SOURCE_CONFLICT_NEEDS_REVIEW"
-    EVIDENCE_CACHE_MISS = "EVIDENCE_CACHE_MISS"
-    EVIDENCE_CACHE_STALE = "EVIDENCE_CACHE_STALE"
-    HUMAN_EVIDENCE_REQUIRED = "HUMAN_EVIDENCE_REQUIRED"
-
-
-SOURCE_CONFIDENCE_BASE: dict[SourceTier, float] = {
-    SourceTier.OFFICIAL_PRIMARY: 0.95,
-    SourceTier.OFFICIAL_SECONDARY: 0.90,
-    SourceTier.AUTHORITATIVE_MEDIA: 0.80,
-    SourceTier.SPECIALIZED_DATABASE: 0.78,
-    SourceTier.GENERAL_WEB: 0.55,
-    SourceTier.USER_SUPPLIED_EVIDENCE: 0.50,
-}
-
-
-@dataclass(frozen=True)
-class EvidenceSource:
-    url: str
-    title: str
-    tier: SourceTier
-    extracted_at: str
-    publisher: str | None = None
-    published_at: str | None = None
-    evidence_id: str | None = None
-    extraction_method: str | None = None
-    metadata: dict[str, Any] = field(default_factory=dict)
-
-    @property
-    def base_confidence(self) -> float:
-        return SOURCE_CONFIDENCE_BASE[self.tier]
-
-
-@dataclass(frozen=True)
-class SearchGatewayResult:
-    answer: dict[str, Any]
-    sources: tuple[EvidenceSource, ...]
-    failure_codes: tuple[FailureCode, ...] = ()
-    fallback_used: bool = False
-    evidence_ids: tuple[str, ...] = ()
-    verified_at: str = field(default_factory=lambda: datetime.now(timezone.utc).isoformat())
-    confidence: float = 0.0
-    confidence_level: str = "low"
-    notes: tuple[str, ...] = ()
-
-
-def confidence_level(score: float) -> str:
-    if score >= 0.85:
-        return "high"
-    if score >= 0.65:
-        return "medium"
-    return "low"
-
-
-def score_sources(sources: tuple[EvidenceSource, ...], failure_codes: tuple[FailureCode, ...] = ()) -> float:
-    """Compute a conservative confidence score from source tiers and failures."""
-    if not sources:
-        return 0.0
-
-    best = max(source.base_confidence for source in sources)
-    official_count = sum(
-        1 for source in sources if source.tier in {SourceTier.OFFICIAL_PRIMARY, SourceTier.OFFICIAL_SECONDARY}
-    )
-    corroboration_bonus = min(0.08, max(0, len(sources) - 1) * 0.03)
-    official_bonus = 0.03 if official_count >= 1 and len(sources) >= 2 else 0.0
-
-    penalty = 0.0
-    if FailureCode.SOURCE_CONFLICT_NEEDS_REVIEW in failure_codes:
-        penalty += 0.25
-    if FailureCode.SEARCH_RESULT_TRUNCATED in failure_codes:
-        penalty += 0.05
-    if FailureCode.SOURCE_EXTRACTION_FAILED in failure_codes:
-        penalty += 0.08
-    if FailureCode.EVIDENCE_CACHE_STALE in failure_codes:
-        penalty += 0.10
-    if FailureCode.HUMAN_EVIDENCE_REQUIRED in failure_codes:
-        penalty += 0.15
-
-    score = best + corroboration_bonus + official_bonus - penalty
-    return max(0.0, min(0.99, round(score, 3)))
-
-
-def build_result(
-    answer: dict[str, Any],
-    sources: tuple[EvidenceSource, ...],
-    failure_codes: tuple[FailureCode, ...] = (),
-    fallback_used: bool = False,
-    notes: tuple[str, ...] = (),
-) -> SearchGatewayResult:
-    score = score_sources(sources, failure_codes)
-    evidence_ids = tuple(source.evidence_id for source in sources if source.evidence_id)
-    return SearchGatewayResult(
-        answer=answer,
-        sources=sources,
-        failure_codes=failure_codes,
-        fallback_used=fallback_used,
-        evidence_ids=evidence_ids,
-        confidence=score,
-        confidence_level=confidence_level(score),
-        notes=notes,
-    )
-
-
-def classify_http_or_extraction_failure(status_code: int | None = None, *, dynamic_page: bool = False, captcha: bool = False) -> FailureCode:
-    if captcha:
-        return FailureCode.SEARCH_PROVIDER_CAPTCHA
-    if status_code == 403:
-        return FailureCode.SOURCE_BLOCKED_BY_CDN
-    if status_code == 404:
-        return FailureCode.SOURCE_404_OR_REMOVED
-    if dynamic_page:
-        return FailureCode.SOURCE_DYNAMIC_PAGE_UNREADABLE
-    return FailureCode.SOURCE_EXTRACTION_FAILED

tests/test_search_gateway_resilience_contract.py

@@ -1,93 +0,0 @@
-from runtime.search_gateway.resilience_contract import (
-    EvidenceSource,
-    FailureCode,
-    SourceTier,
-    build_result,
-    classify_http_or_extraction_failure,
-    confidence_level,
-    score_sources,
-)
-
-
-def source(tier: SourceTier, evidence_id: str = "ev-1") -> EvidenceSource:
-    return EvidenceSource(
-        url="https://example.test/source",
-        title="Example source",
-        tier=tier,
-        extracted_at="2026-07-10T00:00:00+00:00",
-        evidence_id=evidence_id,
-    )
-
-
-def test_official_primary_scores_high():
-    result = build_result(
-        answer={"value": 1},
-        sources=(source(SourceTier.OFFICIAL_PRIMARY),),
-    )
-    assert result.confidence >= 0.85
-    assert result.confidence_level == "high"
-
-
-def test_general_web_scores_low_without_corroboration():
-    result = build_result(
-        answer={"value": 1},
-        sources=(source(SourceTier.GENERAL_WEB),),
-    )
-    assert result.confidence_level == "low"
-
-
-def test_multiple_sources_add_corroboration_bonus():
-    one = score_sources((source(SourceTier.AUTHORITATIVE_MEDIA, "ev-1"),))
-    two = score_sources((source(SourceTier.AUTHORITATIVE_MEDIA, "ev-1"), source(SourceTier.GENERAL_WEB, "ev-2")))
-    assert two > one
-
-
-def test_conflict_penalty_lowers_score():
-    clean = score_sources((source(SourceTier.OFFICIAL_PRIMARY),))
-    conflicted = score_sources((source(SourceTier.OFFICIAL_PRIMARY),), (FailureCode.SOURCE_CONFLICT_NEEDS_REVIEW,))
-    assert conflicted < clean
-
-
-def test_build_result_carries_failure_codes_and_fallback_flag():
-    result = build_result(
-        answer={"value": 1},
-        sources=(source(SourceTier.OFFICIAL_SECONDARY),),
-        failure_codes=(FailureCode.SEARCH_RESULT_TRUNCATED,),
-        fallback_used=True,
-    )
-    assert result.fallback_used is True
-    assert FailureCode.SEARCH_RESULT_TRUNCATED in result.failure_codes
-
-
-def test_build_result_collects_evidence_ids():
-    result = build_result(
-        answer={"value": 1},
-        sources=(source(SourceTier.OFFICIAL_PRIMARY, "ev-a"), source(SourceTier.AUTHORITATIVE_MEDIA, "ev-b")),
-    )
-    assert result.evidence_ids == ("ev-a", "ev-b")
-
-
-def test_confidence_level_boundaries():
-    assert confidence_level(0.85) == "high"
-    assert confidence_level(0.65) == "medium"
-    assert confidence_level(0.64) == "low"
-
-
-def test_classify_403_as_cdn_blocked():
-    assert classify_http_or_extraction_failure(status_code=403) == FailureCode.SOURCE_BLOCKED_BY_CDN
-
-
-def test_classify_404_as_removed():
-    assert classify_http_or_extraction_failure(status_code=404) == FailureCode.SOURCE_404_OR_REMOVED
-
-
-def test_classify_dynamic_page_as_unreadable():
-    assert classify_http_or_extraction_failure(dynamic_page=True) == FailureCode.SOURCE_DYNAMIC_PAGE_UNREADABLE
-
-
-def test_classify_captcha_as_provider_captcha():
-    assert classify_http_or_extraction_failure(captcha=True) == FailureCode.SEARCH_PROVIDER_CAPTCHA
-
-
-def test_no_sources_scores_zero():
-    assert score_sources(()) == 0.0

versions/proposals/V1R4C4-SR02-search-gateway-resilience.json

@@ -1,47 +0,0 @@
-{
-  "sr_id": "V1R4C4-SR02",
-  "title": "Generic Search Gateway Resilience",
-  "baseline": "V1R4C3",
-  "scope": "generic_search_capability",
-  "status": "merge_candidate",
-  "risk": "medium",
-  "owner_gate_required": true,
-  "summary": "Upgrade Search Gateway into a domain-agnostic, evidence-first retrieval pipeline with multi-source fallback, failure diagnosis, evidence cache, and confidence scoring.",
-  "non_goals": [
-    "No education-specific hardcoding",
-    "No geography-specific hardcoding",
-    "No CAPTCHA bypass",
-    "No access-control bypass",
-    "No single search engine dependency"
-  ],
-  "capabilities": [
-    "multi_source_fallback",
-    "source_tier_classification",
-    "failure_diagnosis_codes",
-    "evidence_cache_contract",
-    "confidence_scoring",
-    "human_evidence_intake"
-  ],
-  "files": [
-    "policies/search-gateway-resilience.yaml",
-    "runtime/search_gateway/resilience_contract.py",
-    "tests/test_search_gateway_resilience_contract.py",
-    "docs/search-gateway-resilience.md"
-  ],
-  "acceptance_tests": [
-    "tests/test_search_gateway_resilience_contract.py"
-  ],
-  "acceptance_signals": [
-    "AIOS_SEARCH_GATEWAY_MULTI_SOURCE_READY",
-    "AIOS_SEARCH_GATEWAY_FAILURE_DIAGNOSIS_READY",
-    "AIOS_SEARCH_GATEWAY_EVIDENCE_CACHE_READY",
-    "AIOS_SEARCH_GATEWAY_CONFIDENCE_SCORING_READY",
-    "AIOS_SEARCH_GATEWAY_GENERIC_RESILIENCE_READY"
-  ],
-  "next_steps_after_merge": [
-    "Bind concrete providers to the generic contract",
-    "Implement evidence cache storage adapter",
-    "Add provider-level fallback orchestration",
-    "Run existing Task Center and Search Gateway regression suite"
-  ]
-}