diff --git a/CHANGELOG.md b/CHANGELOG.md
index 8f70ea1b..e8cf664c 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -10,6 +10,27 @@ All notable changes to this project will be documented in this file.
 
 ---
 
+## [0.50.1] - 2026-07-08
+
+### Fixed
+
+- **Release review follow-up**: persist bundle extension payloads through
+  directory save/load round trips so attached requirements evidence survives
+  reload, and make the requirements context adapter protocol stub type-check
+  cleanly.
+
+---
+
+## [0.50.0] - 2026-07-08
+
+### Added
+
+- **Requirements context adapter**: add core helpers for importing,
+  normalizing, validating, and inspecting source-attributed requirement context
+  as validation evidence for the future `requirements` runtime command group.
+
+---
+
 ## [0.49.1] - 2026-07-08
 
 ### Fixed
diff --git a/docs/_data/nav.yml b/docs/_data/nav.yml
index 4b67c57d..94da4b02 100644
--- a/docs/_data/nav.yml
+++ b/docs/_data/nav.yml
@@ -174,6 +174,9 @@
     - title: Requirements Evidence Input Model
       url: /reference/requirements-evidence-input-model/
       expertise: [advanced]
+    - title: Requirements Context Adapter
+      url: /reference/requirements-context-adapter/
+      expertise: [advanced]
     - title: Dependency Resolution
       url: /reference/dependency-resolution/
       expertise: [advanced]
diff --git a/docs/getting-started/installation.md b/docs/getting-started/installation.md
index f83476e1..26b1bb5a 100644
--- a/docs/getting-started/installation.md
+++ b/docs/getting-started/installation.md
@@ -11,7 +11,9 @@ expertise_level: [beginner]
 
 This guide will help you get started with SpecFact CLI in under 60 seconds — first with **no install** (uvx), then with a **persistent** install (pip) when you want IDE workflows and a stable `specfact` command.
 
-> **Primary use case**: brownfield code modernization — reverse-engineering existing codebases into documented specs with runtime contract enforcement. See the [5-Minute Quickstart](quickstart.md) for a full walkthrough.
+> **Primary use case**: validation evidence and AI-bloat defense for AI-assisted and brownfield Python delivery —
+> deterministic review, drift detection, and spec/contract evidence before code reaches PR.
+> See the [5-Minute Quickstart](quickstart.md) for a full walkthrough.
 
 ## Try it now — no install required
 
diff --git a/docs/getting-started/tutorial-openspec-speckit.md b/docs/getting-started/tutorial-openspec-speckit.md
index 1c0ad572..2951875f 100644
--- a/docs/getting-started/tutorial-openspec-speckit.md
+++ b/docs/getting-started/tutorial-openspec-speckit.md
@@ -7,7 +7,7 @@ permalink: /getting-started/tutorial-openspec-speckit/
 # Tutorial: Using SpecFact with OpenSpec or Spec-Kit
 
 > **Complete step-by-step guide for new users**  
-> Learn how to use SpecFact CLI with OpenSpec or Spec-Kit for brownfield code modernization
+> Learn how to use SpecFact CLI with OpenSpec or Spec-Kit to validate existing codebases and keep specs, contracts, and code in sync
 
 **Time**: 15-30 minutes | **Prerequisites**: Python 3.11+, basic command-line knowledge
 
diff --git a/docs/guides/integrations-overview.md b/docs/guides/integrations-overview.md
index b36d6084..8e30d98e 100644
--- a/docs/guides/integrations-overview.md
+++ b/docs/guides/integrations-overview.md
@@ -50,7 +50,8 @@ SpecFact CLI integrations fall into four main categories:
 - Learning and exploration of state machines and contracts
 - Single-developer projects and rapid prototyping
 
-**Key difference**: Spec-Kit focuses on **new feature authoring**, while SpecFact CLI focuses on **brownfield code modernization**.
+**Key difference**: Spec-Kit focuses on **new feature authoring**, while SpecFact CLI focuses on
+**validating existing code and defending it against AI-bloat and drift**.
 
 **See also**: [Spec-Kit Journey Guide](./speckit-journey.md)
 
diff --git a/docs/guides/speckit-journey.md b/docs/guides/speckit-journey.md
index c8c0c5a4..fa0fbc9e 100644
--- a/docs/guides/speckit-journey.md
+++ b/docs/guides/speckit-journey.md
@@ -7,7 +7,7 @@ permalink: /guides/speckit-journey/
 # The Journey: From Spec-Kit to SpecFact
 
 > **Spec-Kit and SpecFact are complementary, not competitive.**  
-> **Primary Use Case**: SpecFact CLI for brownfield code modernization  
+> **Primary Use Case**: SpecFact CLI for validation evidence and AI-bloat defense on existing codebases  
 > **Secondary Use Case**: Add SpecFact enforcement to Spec-Kit's interactive authoring for new features
 
 ---
diff --git a/docs/guides/use-cases.md b/docs/guides/use-cases.md
index 13f32e2d..7289621d 100644
--- a/docs/guides/use-cases.md
+++ b/docs/guides/use-cases.md
@@ -10,7 +10,7 @@ redirect_from:
 
 Detailed use cases and examples for SpecFact CLI.
 
-> **Primary Use Case**: Brownfield code modernization (Use Case 1)  
+> **Primary Use Case**: Validation evidence and AI-bloat defense on existing codebases (Use Case 1)  
 > **Secondary Use Case**: Adding enforcement to Spec-Kit projects (Use Case 2)  
 > **Alternative**: Greenfield spec-first development (Use Case 3)  
 > **Validation Use Case**: Sidecar validation of external codebases (Use Case 4) 🆕
diff --git a/docs/index.md b/docs/index.md
index d5302703..a4a3caf1 100644
--- a/docs/index.md
+++ b/docs/index.md
@@ -10,7 +10,7 @@ doc_owner: specfact-cli
 tracks:
   - src/specfact_cli/**
   - openspec/**
-last_reviewed: 2026-05-21
+last_reviewed: 2026-07-08
 exempt: false
 exempt_reason: ""
 ---
@@ -157,6 +157,7 @@ bundle-specific adapters, and authoring guidance.
 - **[Module Categories](/reference/module-categories/)** - Category taxonomy
 - **[Module Contracts](/reference/module-contracts/)** - Contract interfaces
 - **[Requirements Evidence Input Model](/reference/requirements-evidence-input-model/)** - Requirement input records for validation evidence
+- **[Requirements Context Adapter](/reference/requirements-context-adapter/)** - Import and coverage helpers for requirement context evidence
 - **[Operational Modes](/core-cli/modes/)** - CI/CD and CoPilot modes
 - **[Debug Logging](/core-cli/debug-logging/)** - Diagnostic logging
 
diff --git a/docs/reference/README.md b/docs/reference/README.md
index 821cd44a..eaffd95f 100644
--- a/docs/reference/README.md
+++ b/docs/reference/README.md
@@ -25,6 +25,7 @@ For bundle-specific deep command guides and runbooks, use the canonical modules
 - **[Module Security](module-security.md)** - Marketplace/module integrity and publisher metadata
 - **[Bridge Registry](bridge-registry.md)** - Registry-facing bridge converter declarations
 - **[Requirements Evidence Input Model](requirements-evidence-input-model.md)** - Requirement input records used by validation evidence
+- **[Requirements Context Adapter](requirements-context-adapter.md)** - Import, validation, and coverage helpers for requirement context evidence
 - **[Directory Structure](directory-structure.md)** - Project structure and organization
 - **[Feature Keys](feature-keys.md)** - Key normalization and formats
 - **[Dependency Resolution](dependency-resolution.md)** - Module/pip dependency resolution behavior
diff --git a/docs/reference/directory-structure.md b/docs/reference/directory-structure.md
index 3b439630..aab6fedb 100644
--- a/docs/reference/directory-structure.md
+++ b/docs/reference/directory-structure.md
@@ -20,7 +20,10 @@ Use this document for repository-local artifact placement. Use
 [Module Development](../guides/module-development.md) and
 [Publishing modules](../guides/publishing-modules.md) for source/package layout.
 
-> **Primary Use Case**: SpecFact CLI is designed for **brownfield code modernization** - reverse-engineering existing codebases into documented specs with runtime contract enforcement. The directory structure reflects this brownfield-first approach.
+> **Primary Use Case**: SpecFact CLI produces **validation evidence and AI-bloat defense** for AI-assisted and
+> brownfield Python delivery - reverse-engineering existing codebases into documented specs, then detecting drift,
+> bloat, and weak contracts before merge. The directory structure reflects this evidence-first approach to
+> existing code.
 
 **CLI-First Approach**: SpecFact works offline, requires no account, and integrates with your existing workflow. Works with VS Code, Cursor, GitHub Actions, pre-commit hooks, or any IDE. No platform to learn, no vendor lock-in.
 
diff --git a/docs/reference/module-categories.md b/docs/reference/module-categories.md
index d70486d3..de82cb2d 100644
--- a/docs/reference/module-categories.md
+++ b/docs/reference/module-categories.md
@@ -22,6 +22,7 @@ Category command groups:
 - `specfact code ...`
 - `specfact spec ...`
 - `specfact govern ...`
+- `specfact requirements ...`
 
 ## Canonical Category Assignments
 
@@ -47,6 +48,7 @@ Category command groups:
 | `generate` | `spec` | `specfact-spec` | `spec` | `generate` |
 | `enforce` | `govern` | `specfact-govern` | `govern` | `enforce` |
 | `patch_mode` | `govern` | `specfact-govern` | `govern` | `patch` |
+| `requirements` | `requirements` | `specfact-requirements` | `requirements` | `requirements` |
 
 ## Bundle Contents by Category
 
@@ -55,13 +57,14 @@ Category command groups:
 - `specfact-codebase`: `analyze`, `drift`, `validate`, `repro`
 - `specfact-spec`: `contract`, `api`, `sdd`, `generate`
 - `specfact-govern`: `enforce`, `patch`
+- `specfact-requirements`: `requirements`
 
 ## Bundle Package Layout and Namespaces
 
 Official bundle packages are published from the dedicated modules repository:
 
 - Repository: `nold-ai/specfact-cli-modules`
-- Package roots: `packages/specfact-project/`, `packages/specfact-backlog/`, `packages/specfact-codebase/`, `packages/specfact-spec/`, `packages/specfact-govern/`
+- Package roots: `packages/specfact-project/`, `packages/specfact-backlog/`, `packages/specfact-codebase/`, `packages/specfact-spec/`, `packages/specfact-govern/`, `packages/specfact-requirements/`
 
 Namespace mapping:
 
@@ -70,10 +73,11 @@ Namespace mapping:
 - `specfact-codebase` -> import namespace `specfact_codebase.*`
 - `specfact-spec` -> import namespace `specfact_spec.*`
 - `specfact-govern` -> import namespace `specfact_govern.*`
+- `specfact-requirements` -> import namespace `specfact_requirements.*`
 
 Compatibility note:
 
-- Flat top-level command shims were removed. Use category groups (`project`, `backlog`, `code`, `spec`, `govern`).
+- Flat top-level command shims were removed. Use category groups (`project`, `backlog`, `code`, `spec`, `govern`, `requirements`).
 - `specfact backlog auth ...` is provided by the backlog bundle, not by the permanent core command surface.
 
 > Modules docs handoff: this page remains in the core docs set as release-line overview content.
@@ -115,5 +119,5 @@ Before:
 
 After:
 
-- Core top-level commands plus grouped workflow families (`project`, `backlog`, `code`, `spec`, `govern`).
+- Core top-level commands plus grouped workflow families (`project`, `backlog`, `code`, `spec`, `govern`, `requirements`).
 - No backward-compatibility flat shims.
diff --git a/docs/reference/requirements-context-adapter.md b/docs/reference/requirements-context-adapter.md
new file mode 100644
index 00000000..e0e23495
--- /dev/null
+++ b/docs/reference/requirements-context-adapter.md
@@ -0,0 +1,79 @@
+---
+layout: default
+title: Requirements Context Adapter
+permalink: /reference/requirements-context-adapter/
+description: Core helpers for importing, normalizing, validating, and inspecting upstream requirement context.
+keywords: [requirements, validation, evidence, adapter, projectbundle]
+audience: [team, enterprise]
+expertise_level: [advanced]
+doc_owner: specfact-cli
+tracks:
+  - src/specfact_cli/requirements/context.py
+  - tests/unit/requirements/test_context_adapter.py
+last_reviewed: 2026-07-08
+exempt: false
+exempt_reason: ""
+---
+
+# Requirements Context Adapter
+
+SpecFact core provides helpers that module runtimes can use to normalize
+upstream requirement context into validation evidence. The helpers are designed
+for import, validation, and inspection. They do not create a built-in
+requirements authoring workflow.
+
+## Core Surface
+
+- `normalize_requirement_records(...)` converts source-attributed mappings or
+  `RequirementInput` objects into normalized records and bounded diagnostics.
+- `attach_requirements_to_bundle(...)` stores normalized records under the
+  existing `requirements.inputs` `ProjectBundle` extension.
+- `load_requirements_from_bundle(...)` reads that extension back into
+  `RequirementInput` instances.
+- `validate_requirement_context(...)` emits a `ValidationReport` for evidence
+  usefulness by profile.
+- `inspect_requirement_context_coverage(...)` returns machine-readable coverage
+  counts for downstream command handlers.
+
+```python
+from specfact_cli.models.requirements import RequirementInput, RequirementSourceReference
+from specfact_cli.requirements.context import (
+    attach_requirements_to_bundle,
+    inspect_requirement_context_coverage,
+    normalize_requirement_records,
+)
+
+result = normalize_requirement_records(
+    [
+        {
+            "schema_version": "1",
+            "requirement_id": "REQ-239",
+            "title": "Imported context keeps source attribution",
+            "sources": [
+                {
+                    "source_type": "issue",
+                    "locator": "https://github.com/nold-ai/specfact-cli/issues/239",
+                }
+            ],
+        }
+    ],
+    source_locator="requirements.yaml",
+)
+```
+
+## Runtime Command Ownership
+
+The `requirements` command group is owned by the requirements module runtime.
+Runtime commands should call the core helpers instead of parsing provider
+payloads directly inside root CLI code.
+
+## Compatibility Notes
+
+- Requirement inputs must include `schema_version` and at least one source
+  reference.
+- Invalid imported records produce bounded diagnostics; valid records remain
+  usable.
+- Enterprise or strict validation treats missing downstream evidence links as
+  errors. Less strict profiles receive warnings.
+- Backlog write-back and interactive requirement authoring remain outside this
+  core surface.
diff --git a/docs/reference/requirements-evidence-input-model.md b/docs/reference/requirements-evidence-input-model.md
index a68dec8a..6642e7ad 100644
--- a/docs/reference/requirements-evidence-input-model.md
+++ b/docs/reference/requirements-evidence-input-model.md
@@ -10,7 +10,7 @@ doc_owner: specfact-cli
 tracks:
   - src/specfact_cli/models/requirements.py
   - tests/unit/models/test_requirements.py
-last_reviewed: 2026-07-07
+last_reviewed: 2026-07-08
 exempt: false
 exempt_reason: ""
 ---
@@ -76,3 +76,5 @@ bundle.set_extension(
   validation severity without blocking model construction by themselves.
 - No requirement authoring commands or backlog write-back behavior are provided
   by this model.
+- Import, validation, and coverage helpers for these records live in
+  [Requirements Context Adapter](requirements-context-adapter.md).
diff --git a/openspec/changes/requirements-02-module-commands/CHANGE_VALIDATION.md b/openspec/changes/requirements-02-module-commands/CHANGE_VALIDATION.md
index 4fb21c6d..32ee9e85 100644
--- a/openspec/changes/requirements-02-module-commands/CHANGE_VALIDATION.md
+++ b/openspec/changes/requirements-02-module-commands/CHANGE_VALIDATION.md
@@ -1,31 +1,54 @@
-# Change Validation: requirements-02-module-commands
+# Change Validation Report: requirements-02-module-commands
 
-- **Validated on (UTC):** 2026-02-15T21:54:26Z
-- **Workflow:** /wf-validate-change (proposal-stage dry-run validation)
+- **Validation Date (Europe/Berlin):** 2026-07-08T21:00:03+02:00
+- **Workflow:** OpenSpec validate-change refresh before implementation
 - **Strict command:** `openspec validate requirements-02-module-commands --strict`
 - **Result:** PASS
 
 ## Scope Summary
 
-- **New capabilities:** requirements-module
-- **Modified capabilities:** module-io-contract,backlog-adapter
+- **New capabilities:** requirements-context-adapter
+- **Modified capabilities:** module-io-contract, backlog-adapter
 - **Declared dependencies:** requirements-01 (data model), arch-07 (#213, schema extensions for ProjectBundle)
-- **Proposed affected code paths:** - `modules/requirements/` (new module);- `modules/backlog/src/adapters/` (extend adapters with AC text extraction interface) - `src/specfact_cli/contracts/module_interface.py` (no change — new implementation)
+- **Proposed affected code paths:**
+  - `src/specfact_cli/requirements/context.py`
+  - `src/specfact_cli/requirements/__init__.py`
+  - `tests/unit/requirements/test_context_adapter.py`
+  - `docs/reference/requirements-context-adapter.md`
 
 ## Breaking-Change Analysis (Dry-Run)
 
-- Interface changes are proposal-level only; no production code modifications were performed in this workflow stage.
-- Proposed modified capabilities are additive/extension-oriented in the current spec deltas and do not require immediate breaking migrations at proposal time.
-- Backward-compatibility risk is primarily sequencing-related (dependency ordering), not signature-level breakage at this stage.
+- The change adds new helper APIs for module runtimes to reuse.
+- ProjectBundle integration remains optional through the existing
+  `requirements.inputs` schema-extension namespace.
+- No existing runtime command signature is changed.
+- Existing bundles without `requirements.inputs` remain backward compatible.
 
 ## Dependency and Integration Review
 
-- Dependency declarations align with the 2026-02-15 architecture layer integration plan sequencing.
-- Cross-change integration points are explicitly represented in proposal/spec/task artifacts.
-- No additional mandatory scope expansion was required to pass strict OpenSpec validation.
+- `requirements-01-data-model` (#238) is implemented and archived on
+  2026-07-07.
+- `arch-07-schema-extension-system` (#213) is implemented and archived on
+  2026-02-18.
+- GitHub issue #239 was verified as open and not `in progress` through the
+  GitHub connector on 2026-07-08.
+- The hierarchy cache refresh command succeeded with approved network access and
+  reported no cache changes.
+- The roadmap places #239 under Requirements Layer / Epic #256; the connector
+  does not expose project parent fields, so the local roadmap remains the
+  available structure evidence.
+- The public issue body and title are aligned to the narrowed
+  validation-evidence format.
+- The internal wiki mirror `wiki/sources/requirements-02-module-commands.md`
+  must be updated when scope/status metadata changes and
+  `wiki_rebuild_graph.py` run from the internal repo root.
 
 ## Validation Outcome
 
 - Required artifacts are present: `proposal.md`, `design.md`, `specs/**/*.md`, `tasks.md`.
 - Strict OpenSpec validation passed.
-- Change is ready for implementation-phase intake once prerequisites are satisfied.
+- TDD implementation evidence is complete for the narrowed context-adapter
+  scope, but repository-wide total coverage remains below the 80% policy
+  threshold recorded by CI.
+- Final release readiness is blocked until the required coverage gate is
+  satisfied or an explicit release-governance exception is recorded.
diff --git a/openspec/changes/requirements-02-module-commands/TDD_EVIDENCE.md b/openspec/changes/requirements-02-module-commands/TDD_EVIDENCE.md
new file mode 100644
index 00000000..4122e2ae
--- /dev/null
+++ b/openspec/changes/requirements-02-module-commands/TDD_EVIDENCE.md
@@ -0,0 +1,95 @@
+# TDD Evidence: requirements-02-module-commands
+
+## Failing-before
+
+- **Timestamp (Europe/Berlin):** 2026-07-08T20:48:00+02:00
+- **Command:** `hatch run pytest tests/unit/requirements/test_context_adapter.py -q`
+- **Result:** FAIL, expected
+- **Summary:** Pytest failed during collection because the new core requirements
+  adapter package did not exist yet.
+- **Key error:** `ModuleNotFoundError: No module named 'specfact_cli.requirements'`
+- **Sandbox note:** A first unapproved Hatch run failed before pytest while
+  creating the worktree virtualenv; the approved rerun captured the behavioral
+  missing-package failure above.
+
+## Passing-after
+
+- **Timestamp (Europe/Berlin):** 2026-07-08T20:56:00+02:00
+- **Command:** `hatch run pytest tests/unit/requirements/test_context_adapter.py -q`
+- **Result:** PASS
+- **Summary:** 4 targeted tests passed, covering normalization diagnostics,
+  ProjectBundle extension IO, profile-aware validation severity, machine-readable
+  coverage summaries.
+
+## Quality Gates
+
+- **Final verification timestamp (Europe/Berlin):**
+  2026-07-08T21:26:13+02:00
+- `openspec validate requirements-02-module-commands --strict`: PASS
+- `hatch run format`: PASS, 637 files left unchanged after final edits.
+- `hatch run pytest tests/unit/requirements/test_context_adapter.py -q`: PASS,
+  4 passed.
+- `hatch run type-check`: PASS, 0 errors; repository baseline warnings remain.
+- `hatch run lint`: PASS, pylint 10.00/10.
+- `hatch run yaml-lint`: PASS.
+- `hatch run contract-test`: PASS, cached changed-file contract result.
+- `hatch run smart-test-force`: PASS for test execution, 2,794 tests completed.
+  Total repository coverage reported 64.0%, below the 80% policy threshold;
+  this result is not recorded as a coverage-gate pass.
+  - Test log:
+    `logs/tests/test_run_20260708_211857.log`
+  - Coverage log:
+    `logs/tests/coverage_20260708_211857.log`
+  - Coverage disposition: existing repository-wide coverage gap remains;
+    changed adapter behavior is covered by targeted tests and code review, but
+    total coverage still requires follow-up before the 80% gate can be treated
+    as satisfied.
+  - Note: `hatch run smart-test` used the cached no-relevant-change shortcut
+    after staging and returned nonzero, so the forced full run is the
+    authoritative final suite evidence for this change.
+- `hatch run check-version-sources`: PASS.
+- `hatch run check-pypi-ahead`: PASS, local `0.50.0` ahead of PyPI `0.49.1`.
+- `hatch run verify-modules-signature`: PASS, 4 manifests verified.
+- `hatch run semgrep-sast --json --output /tmp/specfact-req02-semgrep.json`:
+  PASS, 0 findings.
+- `hatch run semgrep-sast-gate --results /tmp/specfact-req02-semgrep.json --baseline tools/semgrep/sast-baseline.json`:
+  PASS.
+- `hatch run bandit-scan`: PASS, no medium/high issues identified.
+- `git diff --check`: PASS.
+- `hatch run specfact code review run --json --out /tmp/req02-code-review-final.json --scope changed`:
+  PASS, no findings on the final code diff.
+- `hatch run specfact code review run --json --out /tmp/req02-code-review-final.json --scope full --path src/specfact_cli/requirements/context.py`:
+  PASS, no findings after the final docs-only scope cleanup.
+
+## PR Review Fix Verification
+
+- **Timestamp (Europe/Berlin):** 2026-07-08T21:54:30+02:00
+- `openspec validate requirements-02-module-commands --strict`: PASS.
+- `hatch run format`: PASS, 637 files left unchanged.
+- `hatch run pytest tests/unit/requirements/test_context_adapter.py -q`:
+  PASS, 5 passed.
+- `hatch run type-check`: PASS, 0 errors; repository baseline warnings remain.
+- `hatch run lint`: PASS, pylint 10.00/10.
+- `hatch run yaml-lint`: PASS.
+- `hatch run python -m crosshair check src/specfact_cli/requirements/context.py --per_condition_timeout=5 --analysis_kind=icontract`:
+  PASS.
+- `hatch run specfact code review run --json --out /tmp/req02-code-review-final.json --scope full --path src/specfact_cli/requirements/context.py`:
+  PASS, no findings.
+- `git diff --check`: PASS.
+
+## CI Compatibility Fix Verification
+
+- **Timestamp (Europe/Berlin):** 2026-07-08T22:07:05+02:00
+- `hatch run pytest tests/unit/registry/test_module_grouping.py tests/unit/requirements/test_context_adapter.py -q`:
+  PASS, 14 passed.
+- `hatch run python` discovery reproduction against fetched
+  `nold-ai/specfact-cli-modules` branch `feature/requirements-02-module-commands`:
+  PASS, discovered 7 packages and normalized `nold-ai/specfact-requirements`
+  to category/group `requirements`.
+- `openspec validate requirements-02-module-commands --strict`: PASS.
+- `hatch run type-check`: PASS, 0 errors; repository baseline warnings remain.
+- `hatch run lint`: PASS, pylint 10.00/10.
+- `hatch run yaml-lint`: PASS.
+- `hatch run specfact code review run --json --out /tmp/req02-code-review-final.json --scope changed`:
+  PASS, no findings.
+- `git diff --check`: PASS.
diff --git a/openspec/changes/requirements-02-module-commands/design.md b/openspec/changes/requirements-02-module-commands/design.md
index 5c0f2309..b4c0afce 100644
--- a/openspec/changes/requirements-02-module-commands/design.md
+++ b/openspec/changes/requirements-02-module-commands/design.md
@@ -1,6 +1,9 @@
 ## Context
 
-This change implements proposal scope for `requirements-02-module-commands` from the 2026-02-15 architecture-layer integration plan. It is proposal-stage only and defines implementation strategy without changing runtime code.
+This change implements the core-owned proposal scope for
+`requirements-02-module-commands` from the 2026-02-15 architecture-layer
+integration plan. Runtime grouped commands remain paired module scope in
+`nold-ai/specfact-cli-modules#165`.
 
 ## Goals / Non-Goals
 
@@ -9,10 +12,14 @@ This change implements proposal scope for `requirements-02-module-commands` from
 - Define an implementation approach that stays within the proposal scope.
 - Keep compatibility with existing module registry, adapter bridge, and contract-first patterns.
 - Preserve offline-first behavior and deterministic CLI execution.
+- Provide core helpers that module runtimes can reuse for import diagnostics,
+  ProjectBundle extension IO, validation reports, and coverage summaries.
 
 **Non-Goals:**
 
-- No production code implementation in this stage.
+- No requirement authoring workflow in this core repository.
+- No runtime `specfact requirements ...` command handlers in this core
+  repository.
 - No schema-breaking changes outside declared capabilities.
 - No dependency expansion beyond the proposal and plan.
 
@@ -22,21 +29,30 @@ This change implements proposal scope for `requirements-02-module-commands` from
 - Keep all public APIs contract-first with `@icontract` and `@beartype`.
 - Make all behavior extensions opt-in or backward-compatible by default.
 - Add/modify OpenSpec deltas first so tests can be derived before implementation.
+- Store normalized requirement context through the existing
+  `requirements.inputs` extension from `requirements-01-data-model`.
+- Treat missing downstream evidence links as profile-sensitive validation
+  findings: strict/enterprise profiles fail, lighter profiles warn.
 
 ## Risks / Trade-offs
 
 - [Dependency ordering drift] -> Mitigation: gate implementation tasks on declared prerequisites.
 - [Capability overlap with adjacent changes] -> Mitigation: keep this change scoped to listed capabilities only.
 - [Documentation drift] -> Mitigation: include explicit docs update tasks in apply phase.
+- [Command ownership confusion] -> Mitigation: core ships reusable helpers only;
+  runtime command handlers stay in modules scope.
 
 ## Migration Plan
 
 1. Implement this change only after listed dependencies are implemented.
 2. Add tests from spec scenarios and capture failing-first evidence.
 3. Implement minimal production changes needed for passing scenarios.
-4. Run quality gates and then open PR to `dev`.
+4. Update public docs, issue body, and internal wiki mirror metadata.
+5. Run quality gates and then open PR to `dev`.
 
 ## Open Questions
 
-- Dependency summary: Depends on requirements-01-data-model and arch-07-schema-extension-system.
-- Whether additional cross-change sequencing constraints should be hard-blocked in `openspec/CHANGE_ORDER.md`.
+- Dependency summary: requirements-01-data-model and
+  arch-07-schema-extension-system are implemented and archived.
+- Runtime module command implementation remains tracked by paired modules issue
+  `nold-ai/specfact-cli-modules#165`.
diff --git a/openspec/changes/requirements-02-module-commands/proposal.md b/openspec/changes/requirements-02-module-commands/proposal.md
index 7708079e..365e2178 100644
--- a/openspec/changes/requirements-02-module-commands/proposal.md
+++ b/openspec/changes/requirements-02-module-commands/proposal.md
@@ -20,23 +20,35 @@ planning source.
 
 ## What Changes
 
-- **NEW**: Import and normalization contract for upstream requirement-like
-  sources, including backlog items, OpenSpec proposals, Spec Kit feature
-  folders, and local markdown/YAML records.
-- **NEW**: Validation command behavior that checks completeness, source
-  freshness, and evidence usefulness by profile.
-- **NEW**: Coverage inspection over normalized inputs, architecture boundaries,
-  contracts, code, tests, and review findings.
-- **NEW**: Adapter hooks that return bounded, source-attributed records rather
-  than free-form planning prose.
+- **NEW**: Core import and normalization helpers for source-attributed
+  requirement records supplied by upstream adapters, including backlog-derived
+  snippets and OpenSpec-derived snippets that already conform to the normalized
+  requirement input contract.
+- **NEW**: Profile-aware validation boundaries that check source attribution and
+  downstream evidence usefulness.
+- **NEW**: Coverage inspection over normalized inputs, architecture, code, test,
+  and validation evidence links.
+- **NEW**: Adapter protocol hooks that return bounded, source-attributed records
+  rather than free-form planning prose.
 - **REMOVED FROM CRITICAL PATH**: Interactive authoring templates and broad
   requirement lifecycle management.
 
+## Out of Scope
+
+- Implementing the runtime `specfact requirements ...` command group in this
+  core repository.
+- Implementing concrete ingestion adapters for Spec Kit feature folders, local
+  markdown files, or YAML requirement records.
+- Interactive requirement authoring as a flagship workflow.
+- Treating `.specfact/requirements/` as the system of record for product
+  management.
+- Bidirectional backlog sync or ceremony automation.
+
 ## Capabilities
 
 ### New Capabilities
 
-- `requirements-validation-commands`: Commands for importing, normalizing,
+- `requirements-context-adapter`: Core helpers for importing, normalizing,
   validating, and inspecting upstream requirement context as validation evidence.
 
 ### Modified Capabilities
@@ -46,6 +58,37 @@ planning source.
 - `backlog-adapter`: Backlog adapters can provide source-attributed requirement
   snippets for validation.
 
+## Impact
+
+- **Affected specs**: `requirements-context-adapter`, `module-io-contract`,
+  `backlog-adapter`
+- **Affected code**:
+  - `src/specfact_cli/requirements/context.py`
+  - `src/specfact_cli/requirements/__init__.py`
+  - `src/specfact_cli/registry/module_grouping.py`
+- **Affected tests**:
+  - `tests/unit/requirements/test_context_adapter.py`
+  - `tests/unit/registry/test_module_grouping.py`
+- **Affected docs**:
+  - `docs/_data/nav.yml`
+  - `docs/index.md`
+  - `docs/reference/README.md`
+  - `docs/reference/module-categories.md`
+  - `docs/reference/requirements-context-adapter.md`
+  - `docs/reference/requirements-evidence-input-model.md`
+  - `CHANGELOG.md`
+- **Integration points**: The paired modules runtime change
+  `nold-ai/specfact-cli-modules#165` consumes these helpers for grouped
+  commands. Core recognizes the `requirements` module category and group command
+  so the paired runtime can mount without shipping root CLI handlers here. This
+  core change does not ship backlog write-back or requirement authoring
+  commands.
+- **Rollback plan**: remove the requirements adapter package, tests, reference
+  page, docs navigation/index updates, requirements evidence model cross-link,
+  OpenSpec change artifacts, and version/changelog entry. Existing
+  `requirements.inputs` bundle data remains compatible because it is still
+  defined by `requirements-01-data-model`.
+
 ---
 
 ## Source Tracking
@@ -55,5 +98,7 @@ planning source.
 - **Issue URL**: <https://github.com/nold-ai/specfact-cli/issues/239>
 - **Paired Modules Runtime Issue**: nold-ai/specfact-cli-modules#165
 - **Paired Modules Scope**: requirements runtime commands
-- **Last Synced Status**: proposed
+- **Repository**: nold-ai/specfact-cli
+- **Last Synced Status**: in_progress
 - **Sanitized**: false
+<!-- content_hash: local-sync-2026-07-08-implementation -->
diff --git a/openspec/changes/requirements-02-module-commands/specs/backlog-adapter/spec.md b/openspec/changes/requirements-02-module-commands/specs/backlog-adapter/spec.md
index f08b2213..65300dd7 100644
--- a/openspec/changes/requirements-02-module-commands/specs/backlog-adapter/spec.md
+++ b/openspec/changes/requirements-02-module-commands/specs/backlog-adapter/spec.md
@@ -2,18 +2,20 @@
 
 ### Requirement: Backlog Adapter
 
-The system SHALL expose backlog acceptance-criteria content to requirements extraction workflows.
+The system SHALL define source-attributed backlog requirement snippets that
+requirements runtime adapters can normalize without provider-specific parsing in
+core command handlers.
 
-#### Scenario: Adapter returns acceptance criteria payload for extraction
+#### Scenario: Adapter returns source fields for requirement context import
 
-- **GIVEN** a backlog item selected for extraction
-- **WHEN** requirements extraction requests source fields
-- **THEN** adapter returns title, description, acceptance-criteria text, and item identity
-- **AND** extraction proceeds without provider-specific parsing in command handlers.
+- **GIVEN** a backlog item selected for requirement context import
+- **WHEN** the requirements adapter receives source fields
+- **THEN** the adapter can return title, description, acceptance-criteria text, and item identity
+- **AND** normalization proceeds without provider-specific parsing in core helpers.
 
 #### Scenario: Missing acceptance criteria is surfaced explicitly
 
 - **GIVEN** a backlog item with no acceptance criteria
-- **WHEN** extraction runs
-- **THEN** item is reported as incomplete input
-- **AND** command output includes the backlog item identifier.
+- **WHEN** requirements context normalization runs
+- **THEN** the item is reported as incomplete input
+- **AND** diagnostics include the backlog item identifier.
diff --git a/openspec/changes/requirements-02-module-commands/specs/module-io-contract/spec.md b/openspec/changes/requirements-02-module-commands/specs/module-io-contract/spec.md
index dbe03fb8..4e023758 100644
--- a/openspec/changes/requirements-02-module-commands/specs/module-io-contract/spec.md
+++ b/openspec/changes/requirements-02-module-commands/specs/module-io-contract/spec.md
@@ -2,18 +2,26 @@
 
 ### Requirement: Module Io Contract
 
-The requirements module SHALL implement all `ModuleIOContract` operations.
+Requirements implementations SHALL consume core requirements context adapter
+helpers through the existing `ModuleIOContract` boundary.
 
 #### Scenario: Import operation maps backlog items to requirements
 
-- **GIVEN** backlog input for import
-- **WHEN** `import_to_bundle` runs
-- **THEN** requirements are added to the bundle with stable IDs
-- **AND** parse diagnostics are included for partial failures.
+- **GIVEN** source-attributed requirement records imported by a module
+- **WHEN** `import_to_bundle` stores normalized requirements on a `ProjectBundle`
+- **THEN** requirements are added under the `requirements.inputs` extension with stable IDs
+- **AND** parse diagnostics remain available to the module runtime for partial failures.
 
 #### Scenario: Validate operation enforces profile schema
 
-- **GIVEN** requirements bundle and active profile schema
-- **WHEN** `validate_bundle` runs
-- **THEN** missing required fields are reported
-- **AND** validation severity respects active policy mode.
+- **GIVEN** a requirements bundle and active validation profile
+- **WHEN** `validate_bundle` delegates to core requirements context validation
+- **THEN** missing evidence links and weak context are reported
+- **AND** validation severity respects the selected evidence strictness profile.
+
+#### Scenario: Requirements runtime mounts through grouped module metadata
+
+- **GIVEN** a paired requirements runtime package with category `requirements`
+- **WHEN** module metadata is discovered by core registry validation
+- **THEN** the `requirements` category is accepted with group command `requirements`
+- **AND** root CLI command handlers remain outside this core change.
diff --git a/openspec/changes/requirements-02-module-commands/specs/requirements-module/spec.md b/openspec/changes/requirements-02-module-commands/specs/requirements-module/spec.md
index 3cd5861a..37c0baf6 100644
--- a/openspec/changes/requirements-02-module-commands/specs/requirements-module/spec.md
+++ b/openspec/changes/requirements-02-module-commands/specs/requirements-module/spec.md
@@ -1,26 +1,28 @@
 ## ADDED Requirements
 
-### Requirement: Requirements Module
+### Requirement: Requirements Context Adapter
 
-The system SHALL provide requirements CLI commands for extract, author, validate, and list.
+The system SHALL provide core requirements context adapter helpers for import,
+normalization, validation, and coverage inspection of upstream requirement
+context as validation evidence.
 
-#### Scenario: Extract command creates requirement artifacts
+#### Scenario: Import helpers normalize source-attributed records
 
-- **GIVEN** `specfact requirements extract --from-backlog <adapter> --output .specfact/requirements/`
-- **WHEN** extraction succeeds
-- **THEN** one or more `*.req.yaml` files are produced
-- **AND** each file includes schema version and source backlog reference.
+- **GIVEN** upstream requirement-like records with source references
+- **WHEN** requirements context normalization runs
+- **THEN** valid records are returned as `RequirementInput` instances
+- **AND** each record keeps schema version and source attribution.
 
-#### Scenario: Author command applies profile-aware template fields
+#### Scenario: Invalid imported records produce bounded diagnostics
 
-- **GIVEN** `specfact requirements author --template story`
-- **WHEN** active profile is `solo`
-- **THEN** authoring prompts require only solo-required fields
-- **AND** optional advanced fields remain non-blocking.
+- **GIVEN** one valid upstream record and one malformed upstream record
+- **WHEN** requirements context normalization runs
+- **THEN** valid records are preserved
+- **AND** the malformed record is reported as a diagnostic without free-form planning prose.
 
-#### Scenario: Validate and list expose completeness and trace coverage
+#### Scenario: Validation and coverage expose evidence usefulness
 
-- **GIVEN** requirement artifacts with trace references
-- **WHEN** `specfact requirements validate` and `specfact requirements list --show-coverage` run
-- **THEN** completeness and coverage are reported per requirement
-- **AND** output is machine-readable when requested.
+- **GIVEN** normalized requirement inputs on a `ProjectBundle`
+- **WHEN** requirements context validation and coverage inspection run
+- **THEN** bundle-level completeness and coverage counts are reported with missing-evidence requirement IDs
+- **AND** the result is machine-readable for downstream module commands.
diff --git a/openspec/changes/requirements-02-module-commands/tasks.md b/openspec/changes/requirements-02-module-commands/tasks.md
index 592c1906..770b01d5 100644
--- a/openspec/changes/requirements-02-module-commands/tasks.md
+++ b/openspec/changes/requirements-02-module-commands/tasks.md
@@ -2,29 +2,37 @@
 
 ## 1. Branch and dependency guardrails
 
-- [ ] 1.1 Create dedicated worktree branch `feature/requirements-02-module-commands` from `dev` before implementation work: `scripts/worktree.sh create feature/requirements-02-module-commands`.
-- [ ] 1.2 Verify prerequisite changes are implemented or explicitly accepted as parallel work.
-- [ ] 1.3 Reconfirm scope against the 2026-02-15 architecture integration plan and this proposal.
+- [x] 1.1 Create dedicated worktree branch `feature/requirements-02-module-commands` from `dev` before implementation work: `scripts/worktree.sh create feature/requirements-02-module-commands`.
+- [x] 1.2 Refresh GitHub hierarchy cache, verify issue #239 is not in progress, and confirm available label/structure metadata.
+- [x] 1.3 Verify prerequisite changes are implemented or explicitly accepted as parallel work.
+- [x] 1.4 Reconfirm scope against `openspec/CHANGE_ORDER.md`: keep this change as import, normalization, validation, and coverage helpers for validation evidence.
+- [x] 1.5 Update the public GitHub issue body to match the narrowed validation-evidence format.
+- [x] 1.6 Update the internal wiki mirror and run `wiki_rebuild_graph.py` from the internal repo root.
 
 ## 2. Spec-first and test-first preparation
 
-- [ ] 2.1 Finalize `specs/` deltas for all listed capabilities and cross-check scenario completeness.
-- [ ] 2.2 Add/update tests mapped to new and modified scenarios.
-- [ ] 2.3 Run targeted tests to capture failing-first behavior and record results in `TDD_EVIDENCE.md`.
+- [x] 2.1 Finalize `specs/` deltas for all listed capabilities and cross-check scenario completeness.
+- [x] 2.2 Add/update tests mapped to new and modified scenarios.
+- [x] 2.3 Run targeted tests to capture failing-first behavior and record results in `TDD_EVIDENCE.md`.
 
 ## 3. Implementation
 
-- [ ] 3.1 Implement minimal production code required to satisfy the new scenarios.
-- [ ] 3.2 Add/update contract decorators and type enforcement on public APIs.
-- [ ] 3.3 Update command wiring, adapters, and models required by this change scope only.
+- [x] 3.1 Implement minimal production code required to satisfy the new scenarios.
+- [x] 3.2 Add/update contract decorators and type enforcement on public APIs.
+- [x] 3.3 Update adapter helpers and models required by this change scope only.
+- [x] 3.4 Keep runtime `specfact requirements ...` command handlers in the paired modules-repo scope; core only exposes reusable helpers.
+- [x] 3.5 Add core registry category compatibility for the paired `requirements` module group without adding root CLI handlers.
 
 ## 4. Validation and documentation
 
-- [ ] 4.1 Re-run tests and quality gates until all changed scenarios pass.
-- [ ] 4.2 Update user-facing docs and navigation for changed/added commands and workflows.
-- [ ] 4.3 Run `openspec validate requirements-02-module-commands --strict` and resolve all issues.
+- [x] 4.1 Re-run tests and quality gates until all changed scenarios pass.
+- [x] 4.2 Update user-facing docs and navigation for changed/added commands and workflows.
+- [x] 4.3 Run module-signature verification; if signed module assets changed, bump module versions and re-sign before PR.
+- [x] 4.4 Run `openspec validate requirements-02-module-commands --strict` and resolve all issues.
+- [x] 4.5 Run SpecFact code review JSON, independent static analysis, and clean-code gates; resolve all findings or document rare explicit exceptions.
 
 ## 5. Delivery
 
-- [ ] 5.1 Update `openspec/CHANGE_ORDER.md` status/dependency notes if implementation sequencing changed.
-- [ ] 5.2 Open a PR from `feature/requirements-02-module-commands` to `dev` with spec/test/code/docs evidence.
+- [x] 5.1 Update version files and `CHANGELOG.md` with a minor feature release entry.
+- [x] 5.2 Review `openspec/CHANGE_ORDER.md` status/dependency notes; update only if implementation sequencing changed.
+- [x] 5.3 Open a PR from `feature/requirements-02-module-commands` to `dev` with spec/test/code/docs evidence.
diff --git a/pyproject.toml b/pyproject.toml
index 9abfbd92..fc5003e5 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "specfact-cli"
-version = "0.49.1"
+version = "0.50.1"
 description = "AI-bloat defense CLI for Python teams. Run deterministic code review, cleanup forecasts, and spec/contract evidence for AI-assisted and brownfield delivery."
 readme = "README.md"
 requires-python = ">=3.11"
diff --git a/setup.py b/setup.py
index 39069804..452e8986 100644
--- a/setup.py
+++ b/setup.py
@@ -7,7 +7,7 @@
 if __name__ == "__main__":
     _setup = setup(
         name="specfact-cli",
-        version="0.49.1",
+        version="0.50.1",
         description=(
             "AI-bloat defense CLI for Python teams. Run deterministic code review, cleanup forecasts, "
             "and spec/contract evidence for AI-assisted and brownfield delivery."
diff --git a/src/__init__.py b/src/__init__.py
index 2d7b5205..a52ea986 100644
--- a/src/__init__.py
+++ b/src/__init__.py
@@ -3,4 +3,4 @@
 """
 
 # Package version: keep in sync with pyproject.toml, setup.py, src/specfact_cli/__init__.py
-__version__ = "0.49.1"
+__version__ = "0.50.1"
diff --git a/src/specfact_cli/__init__.py b/src/specfact_cli/__init__.py
index 95ff063e..ac5607b6 100644
--- a/src/specfact_cli/__init__.py
+++ b/src/specfact_cli/__init__.py
@@ -76,6 +76,6 @@ def _install_progressive_disclosure() -> None:
 # keeps missing-command and missing-parameter UX consistent outside the root CLI too.
 _install_progressive_disclosure()
 
-__version__ = "0.49.1"
+__version__ = "0.50.1"
 
 __all__ = ["__version__"]
diff --git a/src/specfact_cli/models/project.py b/src/specfact_cli/models/project.py
index ce6dd072..19d92dbc 100644
--- a/src/specfact_cli/models/project.py
+++ b/src/specfact_cli/models/project.py
@@ -559,7 +559,13 @@ def load_from_directory(
         if progress_callback:
             progress_callback(current + 1, total_artifacts, "bundle.manifest.yaml")
         manifest_data = load_structured_file(manifest_path)
-        manifest = BundleManifest.model_validate(manifest_data)
+        if not isinstance(manifest_data, dict):
+            raise ValueError("Bundle manifest must be a mapping")
+        manifest_mapping = cast(dict[str, Any], manifest_data)
+        extensions_data = manifest_mapping.get("extensions", {})
+        if not isinstance(extensions_data, dict):
+            raise ValueError("ProjectBundle extensions must be a mapping")
+        manifest = BundleManifest.model_validate(manifest_mapping)
         current += 1
 
         slots = _BundleLoadSlots()
@@ -585,6 +591,7 @@ def load_from_directory(
             features=slots.features,
             clarifications=slots.clarifications,
             change_tracking=change_tracking,
+            extensions=dict(extensions_data),
         )
 
     @beartype
@@ -663,6 +670,8 @@ def save_to_directory(
             progress_callback(total_artifacts, total_artifacts, "bundle.manifest.yaml")
         manifest_path = bundle_dir / "bundle.manifest.yaml"
         manifest_data = self.manifest.model_dump(mode="json")
+        if self.extensions:
+            manifest_data["extensions"] = self.extensions
         if num_features > 1000:
             import json
 
diff --git a/src/specfact_cli/registry/module_grouping.py b/src/specfact_cli/registry/module_grouping.py
index ccc38719..ce4829b1 100644
--- a/src/specfact_cli/registry/module_grouping.py
+++ b/src/specfact_cli/registry/module_grouping.py
@@ -8,17 +8,21 @@
 from specfact_cli.models.module_package import ModulePackageMetadata
 
 
-VALID_CATEGORIES = frozenset({"core", "project", "backlog", "codebase", "spec", "govern"})
+VALID_CATEGORIES = frozenset({"core", "project", "backlog", "codebase", "spec", "govern", "requirements"})
 CATEGORY_TO_GROUP_COMMAND: dict[str, str] = {
     "project": "project",
     "backlog": "backlog",
     "codebase": "code",
     "spec": "spec",
     "govern": "govern",
+    "requirements": "requirements",
 }
 LEGACY_GROUP_COMMAND_ALIASES: dict[tuple[str, str], str] = {
     ("codebase", "codebase"): "code",
 }
+LEGACY_CATEGORY_ALIASES: dict[tuple[str, str], str] = {
+    ("project", "requirements"): "requirements",
+}
 
 
 class ModuleManifestError(Exception):
@@ -71,6 +75,9 @@ def normalize_legacy_bundle_group_command(meta: ModulePackageMetadata) -> Module
     """Normalize known legacy bundle group values to canonical grouped commands."""
     if meta.category is None or meta.bundle_group_command is None:
         return meta
+    normalized_category = LEGACY_CATEGORY_ALIASES.get((meta.category, meta.bundle_group_command))
+    if normalized_category is not None:
+        meta.category = normalized_category
     normalized = LEGACY_GROUP_COMMAND_ALIASES.get((meta.category, meta.bundle_group_command))
     if normalized is not None:
         meta.bundle_group_command = normalized
diff --git a/src/specfact_cli/requirements/__init__.py b/src/specfact_cli/requirements/__init__.py
new file mode 100644
index 00000000..f6ebe080
--- /dev/null
+++ b/src/specfact_cli/requirements/__init__.py
@@ -0,0 +1,28 @@
+"""Requirements context adapter helpers."""
+
+from specfact_cli.requirements.context import (
+    RequirementContextAdapter,
+    RequirementContextCoverageSummary,
+    RequirementContextDiagnostic,
+    RequirementContextDiagnosticSeverity,
+    RequirementContextImportResult,
+    attach_requirements_to_bundle,
+    inspect_requirement_context_coverage,
+    load_requirements_from_bundle,
+    normalize_requirement_records,
+    validate_requirement_context,
+)
+
+
+__all__ = [
+    "RequirementContextAdapter",
+    "RequirementContextCoverageSummary",
+    "RequirementContextDiagnostic",
+    "RequirementContextDiagnosticSeverity",
+    "RequirementContextImportResult",
+    "attach_requirements_to_bundle",
+    "inspect_requirement_context_coverage",
+    "load_requirements_from_bundle",
+    "normalize_requirement_records",
+    "validate_requirement_context",
+]
diff --git a/src/specfact_cli/requirements/context.py b/src/specfact_cli/requirements/context.py
new file mode 100644
index 00000000..8dca5644
--- /dev/null
+++ b/src/specfact_cli/requirements/context.py
@@ -0,0 +1,330 @@
+"""Core requirements context adapter contracts and helpers."""
+
+from __future__ import annotations
+
+from collections.abc import Mapping, Sequence
+from enum import StrEnum
+from pathlib import Path
+from typing import Any, Literal, Protocol, cast, get_args
+
+from beartype import beartype
+from icontract import ensure, require
+from pydantic import BaseModel, ConfigDict, Field, ValidationError
+
+from specfact_cli.models.project import ProjectBundle
+from specfact_cli.models.requirements import (
+    RequirementEvidenceLinkType,
+    RequirementInput,
+    load_requirements_input_extension,
+    requirements_input_extension_payload,
+)
+from specfact_cli.models.validation import ValidationReport
+
+
+RequirementContextValidationProfile = Literal[
+    "solo",
+    "startup",
+    "team",
+    "enterprise",
+    "strict",
+    "solo_developer",
+    "api_first_team",
+    "enterprise_full_stack",
+]
+KNOWN_REQUIREMENT_CONTEXT_PROFILES: frozenset[str] = frozenset(get_args(RequirementContextValidationProfile))
+STRICT_REQUIREMENT_CONTEXT_PROFILES: frozenset[str] = frozenset({"enterprise", "strict", "enterprise_full_stack"})
+
+
+class RequirementContextDiagnosticSeverity(StrEnum):
+    """Severity for bounded requirements context diagnostics."""
+
+    INFO = "info"
+    WARNING = "warning"
+    ERROR = "error"
+
+
+@beartype
+class RequirementContextDiagnostic(BaseModel):
+    """Bounded diagnostic emitted while importing requirement context."""
+
+    model_config = ConfigDict(use_enum_values=True)
+
+    severity: RequirementContextDiagnosticSeverity = Field(..., description="Diagnostic severity")
+    code: str = Field(..., min_length=1, description="Stable diagnostic code")
+    message: str = Field(..., min_length=1, description="Human-readable diagnostic message")
+    source_locator: str | None = Field(default=None, description="Optional upstream source locator")
+    requirement_id: str | None = Field(default=None, description="Best-effort requirement identifier")
+    record_index: int | None = Field(
+        default=None,
+        ge=0,
+        description="Zero-based source record index when available",
+    )
+
+
+@beartype
+class RequirementContextImportResult(BaseModel):
+    """Normalized requirement import result plus bounded diagnostics."""
+
+    requirements: list[RequirementInput] = Field(default_factory=list, description="Normalized requirement records")
+    diagnostics: list[RequirementContextDiagnostic] = Field(
+        default_factory=list,
+        description="Import diagnostics for records that could not be normalized",
+    )
+
+
+@beartype
+class RequirementContextCoverageSummary(BaseModel):
+    """Machine-readable coverage summary for normalized requirement inputs."""
+
+    total_requirements: int = Field(..., ge=0, description="Total normalized requirements")
+    with_business_rules: int = Field(..., ge=0, description="Requirements with business rules")
+    with_constraints: int = Field(..., ge=0, description="Requirements with constraints")
+    with_evidence_links: int = Field(..., ge=0, description="Requirements with evidence links")
+    with_architecture_links: int = Field(..., ge=0, description="Requirements linked to architecture evidence")
+    with_code_links: int = Field(..., ge=0, description="Requirements linked to code evidence")
+    with_test_links: int = Field(..., ge=0, description="Requirements linked to test evidence")
+    missing_evidence_requirement_ids: list[str] = Field(
+        default_factory=list,
+        description="Requirement IDs missing downstream evidence links",
+    )
+
+
+class RequirementContextAdapter(Protocol):
+    """Protocol for module adapters that import requirement-like sources."""
+
+    @beartype
+    @require(lambda source: isinstance(source, Path), "source must be a Path")
+    def import_requirements(self, source: Path, config: Mapping[str, Any]) -> RequirementContextImportResult:
+        """Import source-attributed requirement records from an upstream source."""
+        raise NotImplementedError
+
+
+def _all_records_supported(records: Sequence[RequirementInput | Mapping[str, Any]]) -> bool:
+    return all(isinstance(record, RequirementInput | Mapping) for record in records)
+
+
+def _normalization_result_valid(result: RequirementContextImportResult) -> bool:
+    return all(isinstance(requirement, RequirementInput) for requirement in result.requirements)
+
+
+def _record_requirement_id(record: RequirementInput | Mapping[str, Any]) -> str | None:
+    if isinstance(record, RequirementInput):
+        return _optional_text(record.requirement_id)
+    value = record.get("requirement_id")
+    return _optional_text(value) if isinstance(value, str) and value else None
+
+
+def _validation_error_message(exc: ValidationError) -> str:
+    first_error = exc.errors()[0] if exc.errors() else {"msg": str(exc)}
+    location = ".".join(str(part) for part in first_error.get("loc", ()))
+    message = str(first_error.get("msg", "invalid requirement input")) or "invalid requirement input"
+    return f"{location}: {message}" if location else message
+
+
+def _optional_text(value: str | None) -> str | None:
+    if not value:
+        return None
+    realize = getattr(value, "__ch_realize__", None)
+    if callable(realize):
+        realized = realize()
+        return realized if isinstance(realized, str) else str(realized)
+    return value
+
+
+def _profile_nonempty(profile: str) -> bool:
+    return profile.strip() != ""
+
+
+def _profile_supported(profile: str) -> bool:
+    return profile in KNOWN_REQUIREMENT_CONTEXT_PROFILES
+
+
+def _source_locator_supported(source_locator: str | None) -> bool:
+    return source_locator is None or type(source_locator) is str
+
+
+def _coverage_summary_consistent(result: RequirementContextCoverageSummary) -> bool:
+    return result.total_requirements >= len(result.missing_evidence_requirement_ids)
+
+
+def _requirements_with_evidence_link_type(
+    requirements: Sequence[RequirementInput],
+    link_type: RequirementEvidenceLinkType,
+) -> int:
+    count = 0
+    for requirement in requirements:
+        if any(link.link_type == link_type for link in requirement.evidence_links):
+            count += 1
+    return max(0, count)
+
+
+def _missing_evidence_requirement_ids(requirements: Sequence[RequirementInput]) -> list[str]:
+    return [requirement.requirement_id for requirement in requirements if not requirement.evidence_links]
+
+
+def _invalid_requirement_diagnostic(
+    exc: ValidationError,
+    record: RequirementInput | Mapping[str, Any],
+    source_locator: str | None,
+    record_index: int,
+) -> RequirementContextDiagnostic:
+    return RequirementContextDiagnostic(
+        severity=RequirementContextDiagnosticSeverity.ERROR,
+        code="invalid_requirement_input",
+        message=_validation_error_message(exc),
+        source_locator=_optional_text(source_locator),
+        requirement_id=_record_requirement_id(record),
+        record_index=record_index,
+    )
+
+
+@beartype
+@require(_all_records_supported, "records must be RequirementInput or mapping values")
+@require(_source_locator_supported, "source_locator must be text when provided")
+@ensure(_normalization_result_valid)
+def normalize_requirement_records(
+    records: Sequence[RequirementInput | Mapping[str, Any]],
+    *,
+    source_locator: str | None = None,
+) -> RequirementContextImportResult:
+    """Normalize source-attributed requirement records with bounded diagnostics."""
+    normalized: list[RequirementInput] = []
+    diagnostics: list[RequirementContextDiagnostic] = []
+
+    for record_index, record in enumerate(records):
+        if isinstance(record, RequirementInput):
+            normalized.append(record)
+            continue
+        try:
+            normalized.append(RequirementInput.model_validate(record))
+        except ValidationError as exc:
+            diagnostics.append(_invalid_requirement_diagnostic(exc, record, source_locator, record_index))
+
+    return RequirementContextImportResult(requirements=normalized, diagnostics=diagnostics)
+
+
+@beartype
+@require(lambda bundle: isinstance(bundle, ProjectBundle), "bundle must be a ProjectBundle")
+@require(lambda requirements: all(isinstance(record, RequirementInput) for record in requirements))
+@ensure(lambda result: isinstance(result, ProjectBundle))
+def attach_requirements_to_bundle(bundle: ProjectBundle, requirements: Sequence[RequirementInput]) -> ProjectBundle:
+    """Store normalized requirements on the existing requirements.inputs extension."""
+    bundle.set_extension("requirements", "inputs", requirements_input_extension_payload(list(requirements)))
+    return bundle
+
+
+@beartype
+@require(lambda bundle: isinstance(bundle, ProjectBundle), "bundle must be a ProjectBundle")
+@ensure(lambda result: all(isinstance(record, RequirementInput) for record in result))
+def load_requirements_from_bundle(bundle: ProjectBundle) -> list[RequirementInput]:
+    """Load normalized requirements from the requirements.inputs extension."""
+    payload = bundle.get_extension("requirements", "inputs", default=None)
+    if payload is None:
+        return []
+    if not isinstance(payload, dict):
+        raise ValueError("requirements.inputs extension must be a mapping payload")
+    return load_requirements_input_extension(cast(dict[str, Any], payload))
+
+
+@beartype
+@require(lambda requirements: all(isinstance(record, RequirementInput) for record in requirements))
+@ensure(_coverage_summary_consistent)
+def inspect_requirement_context_coverage(
+    requirements: Sequence[RequirementInput],
+) -> RequirementContextCoverageSummary:
+    """Return machine-readable coverage counts for normalized requirement inputs."""
+    requirement_list = list(requirements)
+    return RequirementContextCoverageSummary(
+        total_requirements=len(requirement_list),
+        with_business_rules=sum(bool(requirement.business_rules) for requirement in requirement_list),
+        with_constraints=sum(bool(requirement.constraints) for requirement in requirement_list),
+        with_evidence_links=sum(bool(requirement.evidence_links) for requirement in requirement_list),
+        with_architecture_links=_requirements_with_evidence_link_type(
+            requirement_list,
+            RequirementEvidenceLinkType.ARCHITECTURE,
+        ),
+        with_code_links=_requirements_with_evidence_link_type(requirement_list, RequirementEvidenceLinkType.CODE),
+        with_test_links=_requirements_with_evidence_link_type(requirement_list, RequirementEvidenceLinkType.TEST),
+        missing_evidence_requirement_ids=_missing_evidence_requirement_ids(requirement_list),
+    )
+
+
+def _missing_evidence_severity(profile: RequirementContextValidationProfile) -> str:
+    return "error" if profile in STRICT_REQUIREMENT_CONTEXT_PROFILES else "warning"
+
+
+def _empty_requirements_report() -> ValidationReport:
+    return ValidationReport(
+        status="warnings",
+        violations=[
+            {
+                "severity": "warning",
+                "message": "No requirements.inputs records are attached to the bundle.",
+                "location": "requirements.inputs",
+            }
+        ],
+        summary={"total_checks": 1, "passed": 0, "failed": 0, "warnings": 1},
+    )
+
+
+def _missing_evidence_violations(
+    coverage: RequirementContextCoverageSummary,
+    *,
+    severity: str,
+) -> list[dict[str, str]]:
+    return [
+        {
+            "severity": severity,
+            "message": "Requirement input has no downstream evidence links.",
+            "location": f"requirements.inputs[{requirement_id}].evidence_links",
+        }
+        for requirement_id in coverage.missing_evidence_requirement_ids
+    ]
+
+
+def _validation_status(failed: int, warnings: int) -> Literal["passed", "failed", "warnings"]:
+    if failed:
+        return "failed"
+    if warnings:
+        return "warnings"
+    return "passed"
+
+
+def _validation_summary(
+    requirements: Sequence[RequirementInput],
+    coverage: RequirementContextCoverageSummary,
+    violations: Sequence[Mapping[str, str]],
+) -> dict[str, int]:
+    failed = sum(violation["severity"] == "error" for violation in violations)
+    warnings = sum(violation["severity"] == "warning" for violation in violations)
+    return {
+        "total_checks": max(1, len(requirements) * 2),
+        "passed": len(requirements) + coverage.with_evidence_links,
+        "failed": failed,
+        "warnings": warnings,
+    }
+
+
+@beartype
+@require(lambda bundle: isinstance(bundle, ProjectBundle), "bundle must be a ProjectBundle")
+@require(_profile_nonempty, "profile must not be empty")
+@require(_profile_supported, "profile must be a supported requirements context validation profile")
+@ensure(lambda result: isinstance(result, ValidationReport))
+def validate_requirement_context(
+    bundle: ProjectBundle,
+    *,
+    profile: RequirementContextValidationProfile = "startup",
+) -> ValidationReport:
+    """Validate requirements context evidence usefulness for a ProjectBundle."""
+    requirements = load_requirements_from_bundle(bundle)
+    if not requirements:
+        return _empty_requirements_report()
+
+    coverage = inspect_requirement_context_coverage(requirements)
+    violations = _missing_evidence_violations(coverage, severity=_missing_evidence_severity(profile))
+    summary = _validation_summary(requirements, coverage, violations)
+    return ValidationReport(
+        status=_validation_status(summary["failed"], summary["warnings"]),
+        violations=violations,
+        summary=summary,
+    )
diff --git a/tests/unit/registry/test_module_grouping.py b/tests/unit/registry/test_module_grouping.py
index 0042046e..c64a7368 100644
--- a/tests/unit/registry/test_module_grouping.py
+++ b/tests/unit/registry/test_module_grouping.py
@@ -58,6 +58,40 @@ def test_module_package_yaml_with_category_codebase_passes_validation(tmp_path:
     assert meta.bundle_sub_command == "analyze"
 
 
+def test_module_package_yaml_with_category_requirements_passes_validation(tmp_path: Path) -> None:
+    """requirements modules mount under the requirements group command."""
+    _write_manifest(
+        tmp_path,
+        "requirements",
+        category="requirements",
+        bundle="specfact-requirements",
+        bundle_group_command="requirements",
+        bundle_sub_command="requirements",
+    )
+    packages = discover_package_metadata(tmp_path, source="builtin")
+    assert len(packages) == 1
+    meta = packages[0][1]
+    assert meta.category == "requirements"
+    assert meta.bundle_group_command == "requirements"
+
+
+def test_prerelease_project_requirements_group_is_normalized(tmp_path: Path) -> None:
+    """Prerelease requirements manifests using project category normalize before validation."""
+    _write_manifest(
+        tmp_path,
+        "requirements",
+        category="project",
+        bundle="specfact-requirements",
+        bundle_group_command="requirements",
+        bundle_sub_command="requirements",
+    )
+    packages = discover_package_metadata(tmp_path, source="builtin")
+    assert len(packages) == 1
+    meta = packages[0][1]
+    assert meta.category == "requirements"
+    assert meta.bundle_group_command == "requirements"
+
+
 def test_legacy_codebase_bundle_group_command_is_normalized(tmp_path: Path) -> None:
     """Legacy marketplace manifests using codebase as group command are normalized to code."""
     _write_manifest(
diff --git a/tests/unit/requirements/__init__.py b/tests/unit/requirements/__init__.py
new file mode 100644
index 00000000..1e12e795
--- /dev/null
+++ b/tests/unit/requirements/__init__.py
@@ -0,0 +1 @@
+"""Requirements context tests."""
diff --git a/tests/unit/requirements/test_context_adapter.py b/tests/unit/requirements/test_context_adapter.py
new file mode 100644
index 00000000..bb7c33df
--- /dev/null
+++ b/tests/unit/requirements/test_context_adapter.py
@@ -0,0 +1,179 @@
+"""Tests for requirements context adapter helpers."""
+
+from __future__ import annotations
+
+import sys
+from pathlib import Path
+from typing import Any, cast
+
+import pytest
+from beartype.roar import BeartypeCallHintParamViolation
+
+from specfact_cli.models.plan import Product
+from specfact_cli.models.project import BundleManifest, BundleVersions, ProjectBundle
+from specfact_cli.models.requirements import (
+    RequirementEvidenceLink,
+    RequirementEvidenceLinkType,
+    RequirementInput,
+    RequirementSourceReference,
+    RequirementSourceType,
+)
+from specfact_cli.requirements.context import (
+    attach_requirements_to_bundle,
+    inspect_requirement_context_coverage,
+    load_requirements_from_bundle,
+    normalize_requirement_records,
+    validate_requirement_context,
+)
+
+
+def _source() -> RequirementSourceReference:
+    return RequirementSourceReference(
+        source_type=RequirementSourceType.ISSUE,
+        locator="https://github.com/nold-ai/specfact-cli/issues/239",
+        title="Requirements Context Adapter Commands",
+    )
+
+
+def _requirement(requirement_id: str = "REQ-239") -> RequirementInput:
+    return RequirementInput(
+        requirement_id=requirement_id,
+        schema_version="1",
+        title="Requirement context is source attributed",
+        sources=[_source()],
+        evidence_links=[
+            RequirementEvidenceLink(
+                link_type=RequirementEvidenceLinkType.TEST,
+                target="tests/unit/requirements/test_context_adapter.py",
+            )
+        ],
+    )
+
+
+def _bundle() -> ProjectBundle:
+    manifest = BundleManifest(
+        versions=BundleVersions(schema="1.0", project="0.1.0"),
+        schema_metadata=None,
+        project_metadata=None,
+    )
+    return ProjectBundle(manifest=manifest, bundle_name="test", product=Product(themes=[], releases=[]))
+
+
+def test_normalize_requirement_records_preserves_valid_records_and_bounds_diagnostics() -> None:
+    """Normalization keeps valid records and reports malformed records deterministically."""
+    result = normalize_requirement_records(
+        [
+            _requirement().model_dump(mode="json"),
+            {"requirement_id": "REQ-BROKEN", "schema_version": "1", "title": "Missing sources"},
+            {"schema_version": "1", "title": "Missing id and sources"},
+        ],
+        source_locator="fixtures/requirements.yaml",
+    )
+
+    assert [record.requirement_id for record in result.requirements] == ["REQ-239"]
+    assert [diagnostic.code for diagnostic in result.diagnostics] == [
+        "invalid_requirement_input",
+        "invalid_requirement_input",
+    ]
+    assert [diagnostic.requirement_id for diagnostic in result.diagnostics] == ["REQ-BROKEN", None]
+    assert [diagnostic.record_index for diagnostic in result.diagnostics] == [1, 2]
+    assert {diagnostic.source_locator for diagnostic in result.diagnostics} == {"fixtures/requirements.yaml"}
+
+
+def test_normalize_requirement_records_preserves_diagnostics_when_crosshair_loaded(
+    monkeypatch: pytest.MonkeyPatch,
+) -> None:
+    """Diagnostic identity fields must not depend on test-tool imports."""
+    monkeypatch.setitem(sys.modules, "crosshair", object())
+
+    result = normalize_requirement_records(
+        [{"requirement_id": "REQ-BROKEN", "schema_version": "1", "title": "Missing sources"}],
+        source_locator="fixtures/requirements.yaml",
+    )
+
+    assert len(result.diagnostics) == 1
+    assert result.diagnostics[0].requirement_id == "REQ-BROKEN"
+    assert result.diagnostics[0].source_locator == "fixtures/requirements.yaml"
+
+
+def test_bundle_helpers_attach_and_load_requirements_input_extension() -> None:
+    """Normalized inputs are stored through the existing ProjectBundle extension namespace."""
+    bundle = _bundle()
+    requirement = _requirement()
+
+    returned = attach_requirements_to_bundle(bundle, [requirement])
+
+    assert returned is bundle
+    loaded = load_requirements_from_bundle(bundle)
+    assert [record.requirement_id for record in loaded] == ["REQ-239"]
+
+
+def test_bundle_helpers_persist_requirements_input_extension(tmp_path: Path) -> None:
+    """Attached requirements survive ProjectBundle directory save/load."""
+    bundle_dir = tmp_path / "test-bundle"
+    bundle = attach_requirements_to_bundle(_bundle(), [_requirement()])
+
+    bundle.save_to_directory(bundle_dir)
+    loaded = ProjectBundle.load_from_directory(bundle_dir)
+
+    requirements = load_requirements_from_bundle(loaded)
+    assert [record.requirement_id for record in requirements] == ["REQ-239"]
+
+
+def test_validate_requirement_context_reports_profile_aware_evidence_gaps() -> None:
+    """Enterprise validation treats missing evidence links as failed evidence usefulness."""
+    requirement = RequirementInput(
+        requirement_id="REQ-NO-EVIDENCE",
+        schema_version="1",
+        title="Requirement lacks downstream evidence",
+        sources=[_source()],
+    )
+    bundle = attach_requirements_to_bundle(_bundle(), [requirement])
+
+    report = validate_requirement_context(bundle, profile="enterprise")
+
+    assert report.status == "failed"
+    assert report.summary == {"total_checks": 2, "passed": 1, "failed": 1, "warnings": 0}
+    assert report.violations[0]["severity"] == "error"
+    assert report.violations[0]["location"] == "requirements.inputs[REQ-NO-EVIDENCE].evidence_links"
+
+
+def test_validate_requirement_context_rejects_unknown_profile() -> None:
+    """Profile selection uses a shared bounded vocabulary to catch typos."""
+    bundle = attach_requirements_to_bundle(_bundle(), [_requirement()])
+
+    with pytest.raises(BeartypeCallHintParamViolation):
+        validate_requirement_context(bundle, profile=cast(Any, "enterprize"))
+
+
+def test_inspect_requirement_context_coverage_is_machine_readable() -> None:
+    """Coverage inspection reports counts downstream command handlers can serialize."""
+    requirement_with_repeated_test_links = _requirement("REQ-WITH-EVIDENCE")
+    requirement_with_repeated_test_links.evidence_links.append(
+        RequirementEvidenceLink(
+            link_type=RequirementEvidenceLinkType.TEST,
+            target="tests/unit/requirements/test_context_adapter.py::duplicate",
+        )
+    )
+    requirements = [
+        requirement_with_repeated_test_links,
+        RequirementInput(
+            requirement_id="REQ-WITHOUT-EVIDENCE",
+            schema_version="1",
+            title="Requirement lacks downstream evidence",
+            sources=[_source()],
+        ),
+    ]
+
+    coverage = inspect_requirement_context_coverage(requirements)
+
+    assert coverage.model_dump() == {
+        "total_requirements": 2,
+        "with_business_rules": 0,
+        "with_constraints": 0,
+        "with_evidence_links": 1,
+        "with_architecture_links": 0,
+        "with_code_links": 0,
+        "with_test_links": 1,
+        "missing_evidence_requirement_ids": ["REQ-WITHOUT-EVIDENCE"],
+    }